document/TranslateProject
2
0
Fork 0
mirror of https://github.com/LCTT/TranslateProject.git synced 2025-01-25 23:11:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

清除部分选题

Browse Source 
@lujun9972 部分选题已经过时
This commit is contained in:
Xingyu Wang 2019-06-09 23:17:07 +08:00
parent c9756c90e4
commit 825f475b5a
23 changed files with 0 additions and 2075 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

116
sources/talk/20171222 18 Cyber-Security Trends Organizations Need to Brace for in 2018.md
View File

@ -1,116 +0,0 @@
18 Cyber-Security Trends Organizations Need to Brace for in 2018
======
### 18 Cyber-Security Trends Organizations Need to Brace for in 2018
Enterprises, end users and governments faced no shortage of security challenges in 2017. Some of those same challenges will continue into 2018, and there will be new problems to solve as well. Ransomware has been a concern for several years and will likely continue to be a big issue in 2018. The new year is also going to bring the formal introduction of the European Union's General Data Protection Regulation (GDPR), which will impact how organizations manage private information. A key trend that emerged in 2017 was an increasing use of artificial intelligence (AI) to help solve cyber-security challenges, and that's a trend that will continue to accelerate in 2018. What else will the new year bring? In this slide show, eWEEK presents 18 security predictions for the year ahead from 18 security experts.
### Africa Emerges as New Area for Threat Actors and Targets
"In 2018, Africa will emerge as a new focus area for cyber-threats--both targeting organizations based there and attacks originating from the continent. With its growth in technology adoption and operations and rising economy, and its increasing number of local resident threat actors, Africa has the largest potential for net-new impactful cyber events." -Steve Stone, IBM X-Force IRIS
### AI vs. AI
"2018 will see a rise in AI-based attacks as cyber-criminals begin using machine learning to spoof human behaviors. The cyber-security industry will need to tune their own AI tools to better combat the new threats. The cat and mouse game of cybercrime and security innovation will rapidly escalate to include AI-enabled tools on both sides." --Caleb Barlow, vice president of Threat Intelligence, IBM Security
### Cyber-Security as a Growth Driver
"CEOs view cyber-security as one of their top risks, but many also see it as an opportunity to innovate and find new ways to generate revenue. In 2018 and beyond, effective cyber-security measures will support companies that are transforming their security, privacy and continuity controls in an effort to grow their businesses." -Greg Bell, KMPG's Global Cyber Security Practice co-leader
### GDPR Means Good Enough Isn't Good Enough
"Too many professionals share a 'good enough' philosophy that they've adopted from their consumer mindset that they can simply upgrade and patch to comply with the latest security and compliance best practices or regulations. In 2018, with the upcoming enforcement of the EU GDPR 'respond fast' rules, organizations will quickly come to terms, and face fines, with why 'good enough' is not 'good' anymore." -Kris Lovejoy, CEO of BluVector
### Consumerization of Cyber-Security
"2018 will mark the debut of the 'consumerization of cyber-security.' This means consumers will be offered a unified, comprehensive suite of security offerings, including, in addition to antivirus and spyware protection, credit and identify abuse monitoring and identity restoration. This is a big step forward compared to what is available in one package today. McAfee Total Protection, which safeguards consumer identities in addition to providing virus and malware protection, is an early, simplified example of this. Consumers want to feel more secure." -Don Dixon, co-founder and managing director, Trident Capital Cybersecurity
### Ransomware Will Continue
"Ransomware will continue to plague organizations with 'old' attacks 'refreshed' and reused. The threat of ransomware will continue into 2018. This year we've seen ransomware wreak havoc across the globe with both WannaCry and NotPetya hitting the headlines. Threats of this type and on this scale will be a common feature of the next 12 months." -Andrew Avanessian, chief operating officer at Avecto
### More Encryption Will Be Needed
"It will become increasingly clear in the industry that HTTPS does not offer the robust security and end-to-end encryption as is commonly believed, and there will be a push to encrypt data before it is sent over HTTPS." -Darren Guccione, CEO and co-founder, Keeper Security
### Denial of Service Will Become Financially Lucrative
"Denial of service will become as financially lucrative as identity theft. Using stolen identities for new account fraud has been the major revenue driver behind breaches. However, in recent years ransomware attacks have caused as much if not more damage, as increased reliance on distributed applications and cloud services results in massive business damage when information, applications or systems are held hostage by attackers." -John Pescatore. SANS' director of emerging security trends
### Goodbye Social Security Number
"2018 is the turning point for the retirement of the Social Security number. At this point, the vast majority of SSNs are compromised, and we can no longer rely on them--nor should we have previously." -Michael Sutton, CISO, Zscaler
### Post-Quantum Cyber-Security Discussion Warms Up the Boardroom
"The uncertainty of cyber-security in a post-quantum world is percolating some circles, but 2018 is the year the discussions gain momentum in the top levels of business. As security experts grapple with preparing for a post-quantum world, top executives will begin to ask what can be done to ensure all of our connected 'things' remain secure." -Malte Pollmann, CEO of Utimaco
### Market Consolidation Is Coming
"There will be accelerated consolidation of cyber niche markets flooded with too many 'me-too' companies offering extremely similar products and services. As an example, authentication, end-point security and threat intelligence now boast a total of more than 25 competitors. Ultimately, only three to six companies in each niche can survive." -Mike Janke, co-founder of DataTribe
### Health Care Will Be a Lucrative Target
"Health records are highly valued on the black market because they are saturated with Personally Identifiable Information (PII). Health care institutions will continue to be a target as they have tighter allocations for security in their IT budgets. Also, medical devices are hard to update and often run on older operating system versions." -Larry Cashdollar, senior engineer, Security Intelligence Response Team, Akamai
### 2018: The Year of Simple Multifactor Authentication for SMBs
"Unfortunately, effective multifactor authentication (MFA) solutions have remained largely out of reach for the average small- and medium-sized business. Though enterprise multifactor technology is quite mature, it often required complex on-premises solutions and expensive hardware tokens that most small businesses couldn't afford or manage. However, the growth of SaaS and smartphones has introduced new multifactor solutions that are inexpensive and easy for small businesses to use. Next year, many SMBs will adopt these new MFA solutions to secure their more privileged accounts and users. 2018 will be the year of MFA for SMBs." -Corey Nachreiner, CTO at WatchGuard Technologies
### Automation Will Improve the IT Skills Gap
"The security skills gap is widening every year, with no signs of slowing down. To combat the skills gap and assist in the growing adoption of advanced analytics, automation will become an even higher priority for CISOs." -Haiyan Song, senior vice president of Security Markets at Splunk
### Industrial Security Gets Overdue Attention
"The high-profile attacks of 2017 acted as a wake-up call, and many plant managers now worry that they could be next. Plant manufacturers themselves will offer enhanced security. Third-party companies going on their own will stay in a niche market. The industrial security manufacturers themselves will drive a cooperation with the security industry to provide security themselves. This is because there is an awareness thing going on and impending government scrutiny. This is different from what happened in the rest of IT/IoT where security vendors just go to market by themselves as a layer on top of IT (i.e.: an antivirus on top of Windows)." -Renaud Deraison, co-founder and CTO, Tenable
### Cryptocurrencies Become the New Playground for Identity Thieves
"The rising value of cryptocurrencies will lead to greater attention from hackers and bad actors. Next year we'll see more fraud, hacks and money laundering take place across the top cryptocurrency marketplaces. This will lead to a greater focus on identity verification and, ultimately, will result in legislation focused on trader identity." -Stephen Maloney, executive vice president of Business Development & Strategy, Acuant
### GDPR Compliance Will Be a Challenge
"In 2018, three quarters of companies or apps will be ruled out of compliance with GDPR and at least one major corporation will be fined to the highest extent in 2018 to set an example for others. Most companies are preparing internally by performing more security assessments and recruiting a mix of security professionals with privacy expertise and lawyers, but with the deadline quickly approaching, it's clear the bulk of businesses are woefully behind and may not be able to avoid these consequences." -Sanjay Beri, founder and CEO, Netskope
### Data Security Solidifies Its Spot in the IT Security Stack
"Many businesses are stuck in the mindset that security of networks, servers and applications is sufficient to protect their data. However, the barrage of breaches in 2017 highlights a clear disconnect between what organizations think is working and what actually works. In 2018, we expect more businesses to implement data security solutions that complement their existing network security deployments." -Jim Varner, CEO of SecurityFirst
### [Eight Cyber-Security Vendors Raise New Funding in November 2017][1]
Though the pace of funding slowed in November, multiple firms raised new venture capital to develop and improve their cyber-security products.
Though the pace of funding slowed in November, multiple firms raised new venture capital to develop and improve their cyber-security products.
--------------------------------------------------------------------------------
via: http://voip.eweek.com/security/18-cyber-security-trends-organizations-need-to-brace-for-in-2018
作者：[Sean Michael Kerner][a]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]:http://voip.eweek.com/Authors/sean-michael-kerner
[1]:http://voip.eweek.com/security/eight-cyber-security-vendors-raise-new-funding-in-november-2017

255
sources/talk/20180209 A review of Virtual Labs virtualization solutions for MOOCs - WebLog Pro Olivier Berger.md
View File

@ -1,255 +0,0 @@
A review of Virtual Labs virtualization solutions for MOOCs – WebLog Pro Olivier Berger
======
### 1 Introduction
This is a memo that tries to capture some of the experience gained in the [FLIRT project][3] on the topic of Virtual Labs for MOOCs (Massive Open Online Courses).
In this memo, we try to draw an overview of some benefits and concerns with existing approaches at using virtualization techniques for running Virtual Labs, as distributions of tools made available for distant learners.
We describe 3 main technical architectures: (1) running Virtual Machine images locally on a virtual machine manager, or (2) displaying the remote execution of similar virtual machines on a IaaS cloud, and (3) the potential of connecting to the remote execution of minimized containers on a remote PaaS cloud.
We then elaborate on some perspectives for locally running ports of applications to the WebAssembly virtual machine of the modern Web browsers.
Disclaimer: This memo doesn’t intend to point to extensive literature on the subject, so part of our analysis may be biased by our particular context.
### 2 Context : MOOCs
Many MOOCs (Massive Open Online Courses) include a kind of “virtual laboratory” for learners to experiment with tools, as a way to apply the knowledge, practice, and be more active in the learning process. In quite a few (technical) disciplines, this can consist in using a set of standard applications in a professional domain, which represent typical tools that would be used in real life scenarii.
Our main perspective will be that of a MOOC editor and of MOOC production teams which want to make “virtual labs” available for MOOC participants.
Such a “virtual lab” would typically contain installations of existing applications, pre-installed and configured, and loaded with scenario data in order to perform a lab.
The main constraint here is that such labs would typically be fabricated with limited software development expertise and funds[1][4]. Thus we consider here only the assembly of existing “normal” applications and discard the option of developping novel “serious games” and simulator applications for such MOOCs.
#### 2.1 The FLIRT project
The [FLIRT project][5] groups a consortium of 19 partners in Industry, SMEs and Academia to work on a collection of MOOCs and SPOCs for professional development in Networks and Telecommunications. Lead by Institut Mines Telecom, it benefits from the funding support of the French “Investissements d’avenir” programme.
As part of the FLIRT roadmap, we’re leading an “innovation task” focused on Virtual Labs in the context of the Cloud. This memo was produced as part of this task.
#### 2.2 Some challenges in virtual labs design for distant learning
Virtual Labs used in distance learning contexts require the use of software applications in autonomy, either running on a personal, or professional computer. In general, the technical skills of participants may be diverse. So much for the quality (bandwith, QoS, filtering, limitations: firewaling) of the hardware and networks they use at home or at work. It’s thus very optimistic to seek for one solution fits all strategy.
Most of the time there’s a learning curve on getting familiar with the tools which students will have to use, which constitutes as many challenges to overcome for beginners. These tools may not be suited for beginners, but they will still be selected by the trainers as they’re representative of the professional context being taught.
In theory, this usability challenge should be addressed by devising an adapted pedagogical approach, especially in a context of distance learning, so that learners can practice the labs on their own, without the presence of a tutor or professor. Or some particular prerequisite skills could be required (“please follow System Administration 101 before applying to this course”).
Unfortunately there are many cases where instructors basically just translate to a distant learning scenario, previous lab resources that had previously been devised for in presence learning. This lets learner faced with many challenges to overcome. The only support resource is often a regular forum on the MOOC’s LMS (Learning Management System).
My intuition[2][6] is that developing ad-hoc simulators for distant education would probably be more efficient and easy to use for learners. But that would require a too high investment for the designers of the courses.
In the context of MOOCs which are mainly free to participate to, not much investment is possible in devising ad-hoc lab applications, and instructors have to rely on existing applications, tools and scenarii to deliver a cheap enough environment. Furthermore, technical or licensing constraints[3][7] may lead to selecting lab tools which may not be easy to learn, but have the great advantage or being freely redistributable[4][8].
### 3 Virtual Machines for Virtual Labs
The learners who will try unattended learning in such typical virtual labs will face difficulties in making specialized applications run. They must overcome the technical details of downloading, installing and configuring programs, before even trying to perform a particular pedagogical scenario linked to the matter studied.
To diminish these difficulties, one traditional approach for implementing labs in MOOCs has been to assemble in advance a Virtual Machine image. This already made image can then be downloaded and run with a virtual machine simulator (like [VirtualBox][9][5][10]).
The pre-loaded VM will already have everything ready for use, so that the learners don’t have to install anything on their machines.
An alternative is to let learners download and install the needed software tools themselves, but this leads to so many compatibility issues or technical skill prerequisites, that this is often not advised, and mentioned only as a fallback option.
#### 3.1 Downloading and installation issues
Experience shows[2][11] that such virtual machines also bring some issues. Even if installation of every piece of software is no longer required, learners still need to be able to run the VM simulator on a wide range of diverse hardware, OSes and configurations. Even managing to download the VMs, still causes many issues (lack admin privileges, weight vs download speed, memory or CPU load, disk space, screen configurations, firewall filtering, keayboard layout, etc.).
These problems aren’t generally faced by the majority of learners, but the impacted minority is not marginal either, and they generally will produce a lot of support requests for the MOOC team (usually in the forums), which needs to be anticipated by the community managers.
The use of VMs is no show stopper for most, but can be a serious problem for a minority of learners, and is then no silver bullet.
Some general usability issues may also emerge if users aren’t used to the look and feel of the enclosed desktop. For instance, the VM may consist of a GNU/Linux desktop, whereas users would use a Windows or Mac OS system.
#### 3.2 Fabrication issues for the VM images
On the MOOC team’s side, the fabrication of a lightweight, fast, tested, license-free and easy to use VM image isn’t necessarily easy.
Software configurations tend to rot as time passes, and maintenance may not be easy when the later MOOC editions evolutions lead to the need to maintain the virtual lab scenarii years later.
Ideally, this would require adopting an “industrial” process in building (and testing) the lab VMs, but this requires quite an expertise (system administration, packaging, etc.) that may or not have been anticipated at the time of building the MOOC (unlike video editing competence, for instance).
Our experiment with the [Vagrant][12] technology [[0][13]] and Debian packaging was interesting in this respect, as it allowed us to use a well managed “script” to precisely control the build of a minimal VM image.
### 4 Virtual Labs as a Service
To overcome the difficulties in downloading and running Virtual Machines on one’s local computer, we have started exploring the possibility to run these applications in a kind of Software as a Service (SaaS) context, “on the cloud”.
But not all applications typically used in MOOC labs are already available for remote execution on the cloud (unless the course deals precisely with managing email in GMail).
We have then studied the option to use such an approach not for a single application, but for a whole virtual “desktop” which would be available on the cloud.
#### 4.1 IaaS deployments
A way to achieve this goal is to deploy Virtual Machine images quite similar to the ones described above, on the cloud, in an Infrastructure as a Service (IaaS) context[6][14], to offer access to remote desktops for every learners.
There are different technical options to achieve this goal, but a simplified description of the architecture can be seen as just running Virtual Machines on a single IaaS platform instead of on each learner’s computer. Access to the desktop and application interfaces is made possible with the use of Web pages (or other dedicated lightweight clients) which will display a “full screen” display of the remote desktop running for the user on the cloud VM. Under the hood, the remote display of a Linux desktop session is made with technologies like [VNC][15] and [RDP][16] connecting to a [Guacamole][17] server on the remote VM.
In the context of the FLIRT project, we have made early experiments with such an architecture. We used the CloVER solution by our partner [ProCAN][18] which provides a virtual desktops broker between [OpenEdX][19] and an [OpenStack][20] IaaS public platform.
The expected benefit is that users don’t have to install anything locally, as the only tool needed locally is a Web browser (displaying a full-screen [HTML5 canvas][21] displaying the remote desktop run by the Guacamole server running on the cloud VM.
But there are still some issues with such an approach. First, the cost of operating such an infrastructure : Virtual Machines need to be hosted on a IaaS platform, and that cost of operation isn’t null[7][22] for the MOOC editor, compared to the cost of VirtualBox and a VM running on the learner’s side (basically zero for the MOOC editor).
Another issue, which could be more problematic lies in the need for a reliable connection to the Internet during the whole sequences of lab execution by the learners[8][23]. Even if Guacamole is quite efficient at compressing rendering traffic, some basic connectivity is needed during the whole Lab work sessions, preventing some mobile uses for instance.
One other potential annoyance is the potential delays for making a VM available to a learner (provisioning a VM), when huge VMs images need to be copied inside the IaaS platform when a learner connects to the Virtual Lab activity for the first time (several minutes delays). This may be worse if the VM image is too big (hence the need for optimization of the content[9][24]).
However, the fact that all VMs are running on a platform under the control of the MOOC editor allows new kind of features for the MOOC. For instance, learners can submit results of their labs directly to the LMS without the need to upload or copy-paste results manually. This can help monitor progress or perform evaluation or grading.
The fact that their VMs run on the same platform also allows new kinds of pedagogical scenarii, as VMs of multiple learners can be interconnected, allowing cooperative activities between learners. The VM images may then need to be instrumented and deployed in particular configurations, which may require the use of a dedicated broker like CloVER to manage such scenarii.
For the records, we have yet to perform a rigorous benchmarking of such a solution in order to evaluate its benefits, or constraints given particular contexts. In FLIRT, our main focus will be in the context of SPOCs for professional training (a bit different a context than public MOOCs).
Still this approach doesn’t solve the VMs fabrication issues for the MOOC staff. Installing software inside a VM, be it local inside a VirtualBox simulator of over the cloud through a remote desktop display, makes not much difference. This relies mainly on manual operations and may not be well managed in terms of quality of the process (reproducibility, optimization).
#### 4.2 PaaS deployments using containers
Some key issues in the IaaS context described above, are the cost of operation of running full VMs, and long provisioning delays.
We’re experimenting with new options to address these issues, through the use of [Linux containers][25] running on a PaaS (Platform as a Service) platform, instead of full-fleshed Virtual Machines[10][26].
The main difference, with containers instead of Virtual Machines, lies in the reduced size of images, and much lower CPU load requirements, as the container remove the need for one layer of virtualization. Also, the deduplication techniques at the heart of some virtual file-systems used by container platforms lead to really fast provisioning, avoiding the need to wait for the labs to start.
The traditional making of VMs, done by installing packages and taking a snapshot, was affordable for the regular teacher, but involved manual operations. In this respect, one other major benefit of containers is the potential for better industrialization of the virtual lab fabrication, as they are generally not assembled manually. Instead, one uses a “scripting” approach in describing which applications and their dependencies need to be put inside a container image. But this requires new competence from the Lab creators, like learning the [Docker][27] technology (and the [OpenShift][28] PaaS, for instance), which may be quite specialized. Whereas Docker containers tend to become quite popular in Software Development faculty (through the “[devops][29]” hype), they may be a bit new to other field instructors.
The learning curve to mastering the automation of the whole container-based labs installation needs to be evaluated. There’s a trade-off to consider in adopting technology like Vagrant or Docker: acquiring container/PaaS expertise vs quality of industrialization and optimization. The production of a MOOC should then require careful planning if one has to hire or contract with a PaaS expert for setting up the Virtual Labs.
We may also expect interesting pedagogical benefits. As containers are lightweight, and platforms allow to “easily” deploy multiple interlinked containers (over dedicated virtual networks), this enables the setup of more realistic scenarii, where each learner may be provided with multiple “nodes” over virtual networks (all running their individual containers). This would be particularly interesting for Computer Networks or Security teaching for instance, where each learner may have access both to client and server nodes, to study client-server protocols, for instance. This is particularly interesting for us in the context of our FLIRT project, where we produce a collection of Computer Networks courses.
Still, this mode of operation relies on a good connectivity of the learners to the Cloud. In contexts of distance learning in poorly connected contexts, the PaaS architecture doesn’t solve that particular issue compared to the previous IaaS architecture.
### 5 Future server-less Virtual Labs with WebAssembly
As we have seen, the IaaS or PaaS based Virtual Labs running on the Cloud offer alternatives to installing local virtual machines on the learner’s computer. But they both require to be connected for the whole duration of the Lab, as the applications would be executed on the remote servers, on the Cloud (either inside VMs or containers).
We have been thinking of another alternative which could allow the deployment of some Virtual Labs on the local computers of the learners without the hassles of downloading and installing a Virtual Machine manager and VM image. We envision the possibility to use the infrastructure provided by modern Web browsers to allow running the lab’s applications.
At the time of writing, this architecture is still highly experimental. The main idea is to rebuild the applications needed for the Lab so that they can be run in the “generic” virtual machine present in the modern browsers, the [WebAssembly][30] and Javascript execution engine.
WebAssembly is a modern language which seeks for maximum portability, and as its name hints, is a kind of assembly language for the Web platform. What is of interest for us is that WebAssembly is portable on most modern Web browsers, making it a very interesting platform for portability.
Emerging toolchains allow recompiling applications written in languages like C or C++ so that they can be run on the WebAssembly virtual machine in the browser. This is interesting as it doesn’t require modifying the source code of these programs. Of course, there are limitations, in the kind of underlying APIs and libraries compatible with that platform, and on the sandboxing of the WebAssembly execution engine enforced by the Web browser.
Historically, WebAssembly has been developped so as to allow running games written in C++ for a framework like Unity, in the Web browser.
In some contexts, for instance for tools with an interactive GUI, and processing data retrieved from files, and which don’t need very specific interaction with the underlying operating system, it seems possible to port these programs to WebAssembly for running them inside the Web browser.
We have to experiment deeper with this technology to validate its potential for running Virtual Labs in the context of a Web browser.
We used a similar approach in the past in porting a Relational Database course lab to the Web browser, for standalone execution. A real database would run in the minimal SQLite RDBMS, recompiled to JavaScript[11][31]. Instead of having to download, install and run a VM with a RDBMS, the students would only connect to a Web page, which would load the DBMS in memory, and allow performing the lab SQL queries locally, disconnected from any third party server.
In a similar manner, we can think for instance, of a Lab scenario where the Internet packet inspector features of the Wireshark tool would run inside the WebAssembly virtual machine, to allow dissecting provided capture files, without having to install Wireshard, directly into the Web browser.
We expect to publish a report on that last experiment in the future with more details and results.
### 6 Conclusion
The most promising architecture for Virtual Lab deployments seems to be the use of containers on a PaaS platform for deploying virtual desktops or virtual application GUIs available in the Web browser.
This would allow the controlled fabrication of Virtual Labs containing the exact bits needed for learners to practice while minimizing the delays.
Still the need for always-on connectivity can be a problem.
Also, the potential for inter-networked containers allowing the kind of multiple nodes and collaborative scenarii we described, would require a lot of expertise to develop, and management platforms for the MOOC operators, which aren’t yet mature.
We hope to be able to report on our progress in the coming months and years on those aspects.
### 7 References
[0]
Olivier Berger, J Paul Gibson, Claire Lecocq and Christian Bac “Designing a virtual laboratory for a relational database MOOC”. International Conference on Computer Supported Education, SCITEPRESS, 23-25 may 2015, Lisbonne, Portugal, 2015, vol. 7, pp. 260-268, ISBN 978-989-758-107-6 – [DOI: 10.5220/0005439702600268][1] ([preprint (HTML)][2])
### 8 Copyright
[![Creative Commons License](https://i.creativecommons.org/l/by-nc-sa/4.0/88x31.png)][45]
This work is licensed under a [Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License][46]
.
### Footnotes:
[1][32] – The FLIRT project also works on business model aspects of MOOC or SPOC production in the context of professional development, but the present memo starts from a minimalitic hypothesis where funding for course production is quite limited.
[2][33] – research-based evidence needed
[3][34] – In typical MOOCs which are free to participate, the VM should include only gratis tools, which typically means a GNU/Linux distribution loaded with applications available under free and open source licenses.
[4][35] – Typically, Free and Open Source software, aka Libre Software
[5][36] – VirtualBox is portable on many operating systems, making it a very popular solution for such a need
[6][37] – the IaaS platform could typically be an open cloud for MOOCs or a private cloud for SPOCs (for closer monitoring of student activity or security control reasons).
[7][38] – Depending of the expected use of the lab by learners, this cost may vary a lot. The size and configuration required for the included software may have an impact (hence the need to minimize the footprint of the VM images). With diminishing costs in general this may not be a show stopper. Refer to marketing figures of commercial IaaS offerings for accurate figures. Attention to additional licensing costs if the OS of the VM isn’t free software, or if other licenses must be provided for every learners.
[8][39] – The needs for always-on connectivity may not be a problem for professional development SPOCs where learners connect from enterprise networks for instance. It may be detrimental when MOOCs are very popular in southern countries where high bandwidth is both unreliable and expensive.
[9][40] – In this respect, providing a full Linux desktop inside the VM doesn’t necessarily make sense. Instead, running applications full-screen may be better, avoiding installation of whole desktop environments like Gnome or XFCE… but which has usability consequences. Careful tuning and testing is needed in any case.
[10][41] – The availability of container based architectures is quite popular in the industry, but has not yet been deployed to a large scale in the context of large public MOOC hosting platforms, to our knowledge, at the time of writing. There are interesting technical challenges which the FLIRT project tries to tackle together with its partner ProCAN.
[11][42] – See the corresponding paragraph [http://www-inf.it-sudparis.eu/PROSE/csedu2015/#standalone-sql-env][43] in [0][44]
--------------------------------------------------------------------------------
via: https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/
作者：[Author;Olivier Berger;Télécom Sudparis][a]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]:https://www-public.tem-tsp.eu
[1]:http://dx.doi.org/10.5220/0005439702600268
[2]:http://www-inf.it-sudparis.eu/PROSE/csedu2015/
[3]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#org50fdc1a
[4]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.1
[5]:http://flirtmooc.wixsite.com/flirt-mooc-telecom
[6]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.2
[7]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.3
[8]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.4
[9]:http://virtualbox.org
[10]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.5
[11]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.2
[12]:https://www.vagrantup.com/
[13]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#orgde5af50
[14]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.6
[15]:https://en.wikipedia.org/wiki/Virtual_Network_Computing
[16]:https://en.wikipedia.org/wiki/Remote_Desktop_Protocol
[17]:http://guacamole.apache.org/
[18]:https://www.procan-group.com/
[19]:https://open.edx.org/
[20]:http://openstack.org/
[21]:https://en.wikipedia.org/wiki/Canvas_element
[22]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.7
[23]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.8
[24]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.9
[25]:https://www.redhat.com/en/topics/containers
[26]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.10
[27]:https://en.wikipedia.org/wiki/Docker_(software)
[28]:https://www.openshift.com/
[29]:https://en.wikipedia.org/wiki/DevOps
[30]:http://webassembly.org/
[31]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fn.11
[32]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.1
[33]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.2
[34]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.3
[35]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.4
[36]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.5
[37]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.6
[38]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.7
[39]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.8
[40]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.9
[41]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.10
[42]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#fnr.11
[43]:http://www-inf.it-sudparis.eu/PROSE/csedu2015/#standalone-sql-env
[44]:https://www-public.tem-tsp.eu/~berger_o/weblog/a-review-of-virtual-labs-virtualization-solutions-for-moocs/#orgde5af50
[45]:http://creativecommons.org/licenses/by-nc-sa/4.0/
[46]:http://creativecommons.org/licenses/by-nc-sa/4.0/

45
sources/talk/20180705 New Training Options Address Demand for Blockchain Skills.md
View File

@ -1,45 +0,0 @@
New Training Options Address Demand for Blockchain Skills
======
![](https://www.linux.com/sites/lcom/files/styles/rendered_file/public/blockchain-301.png?itok=1EA-Ob6F)
Blockchain technology is transforming industries and bringing new levels of trust to contracts, payment processing, asset protection, and supply chain management. Blockchain-related jobs are the second-fastest growing in today’s labor market, [according to TechCrunch][1]. But, as in the rapidly expanding field of artificial intelligence, there is a pronounced blockchain skills gap and a need for expert training resources.
### Blockchain for Business
A new training option was recently announced from The Linux Foundation. Enrollment is now open for a free training course called[Blockchain: Understanding Its Uses and Implications][2], as well as a [Blockchain for Business][2] professional certificate program. Delivered through the edX training platform, the new course and program provide a way to learn about the impact of blockchain technologies and a means to demonstrate that knowledge. Certification, in particular, can make a difference for anyone looking to work in the blockchain arena.
“In the span of only a year or two, blockchain has gone from something seen only as related to cryptocurrencies to a necessity for businesses across a wide variety of industries,” [said][3] Linux Foundation General Manager, Training & Certification Clyde Seepersad. “Providing a free introductory course designed not only for technical staff but business professionals will help improve understanding of this important technology, while offering a certificate program through edX will enable professionals from all over the world to clearly demonstrate their expertise.”
TechCrunch [also reports][4] that venture capital is rapidly flowing toward blockchain-focused startups. And, this new program is designed for business professionals who need to understand the potential – or threat – of blockchain to their company and industry.
“Professional Certificate programs on edX deliver career-relevant education in a flexible, affordable way, by focusing on the critical skills industry leaders and successful professionals are seeking today,” said Anant Agarwal, edX CEO and MIT Professor.
### Hyperledger Fabric
The Linux Foundation is steward to many valuable blockchain resources and includes some notable community members. In fact, a recent New York Times article — “[The People Leading the Blockchain Revolution][5]” — named Brian Behlendorf, Executive Director of The Linux Foundation’s [Hyperledger Project][6], one of the [top influential voices][7] in the blockchain world.
Hyperledger offers proven paths for gaining credibility and skills in the blockchain space. For example, the project offers a free course titled Introduction to Hyperledger Fabric for Developers. Fabric has emerged as a key open source toolset in the blockchain world. Through the Hyperledger project, you can also take the B9-lab Certified Hyperledger Fabric Developer course. More information on both courses is available [here][8].
“As you can imagine, someone needs to do the actual coding when companies move to experiment and replace their legacy systems with blockchain implementations,” states the Hyperledger website. “With training, you could gain serious first-mover advantage.”
--------------------------------------------------------------------------------
via: https://www.linux.com/blog/2018/7/new-training-options-address-demand-blockchain-skills
作者：[SAM DEAN][a]
选题：[lujun9972](https://github.com/lujun9972)
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]:https://www.linux.com/users/sam-dean
[1]:https://techcrunch.com/2018/02/14/blockchain-engineers-are-in-demand/
[2]:https://www.edx.org/course/understanding-blockchain-and-its-implications
[3]:https://www.linuxfoundation.org/press-release/as-demand-skyrockets-for-blockchain-expertise-the-linux-foundation-and-edx-offer-new-introductory-blockchain-course-and-blockchain-for-business-professional-certificate-program/
[4]:https://techcrunch.com/2018/05/20/with-at-least-1-3-billion-invested-globally-in-2018-vc-funding-for-blockchain-blows-past-2017-totals/
[5]:https://www.nytimes.com/2018/06/27/business/dealbook/blockchain-stars.html
[6]:https://www.hyperledger.org/
[7]:https://www.linuxfoundation.org/blog/hyperledgers-brian-behlendorf-named-as-top-blockchain-influencer-by-new-york-times/
[8]:https://www.hyperledger.org/resources/training

185
sources/talk/20180802 How blockchain will influence open source.md
View File

@ -1,185 +0,0 @@
How blockchain will influence open source
======
![](https://opensource.com/sites/default/files/styles/image-full-size/public/lead-images/block-quilt-chain.png?itok=mECoDbrc)
What [Satoshi Nakamoto][1] started as Bitcoin a decade ago has found a lot of followers and turned into a movement for decentralization. For some, blockchain technology is a religion that will have the same impact on humanity as the Internet has had. For others, it is hype and technology suitable only for Ponzi schemes. While blockchain is still evolving and trying to find its place, one thing is for sure: It is a disruptive technology that will fundamentally transform certain industries. And I'm betting open source will be one of them.
### The open source model
Open source is a collaborative software development and distribution model that allows people with common interests to gather and produce something that no individual can create on their own. It allows the creation of value that is bigger than the sum of its parts. Open source is enabled by distributed collaboration tools (IRC, email, git, wiki, issue trackers, etc.), distributed and protected by an open source licensing model and often governed by software foundations such as the [Apache Software Foundation][2] (ASF), [Cloud Native Computing Foundation][3] (CNCF), etc.
One interesting aspect of the open source model is the lack of financial incentives in its core. Some people believe open source work should remain detached from money and remain a free and voluntary activity driven only by intrinsic motivators (such as "common purpose" and "for the greater good”). Others believe open source work should be rewarded directly or indirectly through extrinsic motivators (such as financial incentive). While the idea of open source projects prospering only through voluntary contributions is romantic, in reality, the majority of open source contributions are done through paid development. Yes, we have a lot of voluntary contributions, but these are on a temporary basis from contributors who come and go, or for exceptionally popular projects while they are at their peak. Creating and sustaining open source projects that are useful for enterprises requires developing, documenting, testing, and bug-fixing for prolonged periods, even when the software is no longer shiny and exciting. It is a boring activity that is best motivated through financial incentives.
### Commercial open source
Software foundations such as ASF survive on donations and other income streams such as sponsorships, conference fees, etc. But those funds are primarily used to run the foundations, to provide legal protection for the projects, and to ensure there are enough servers to run builds, issue trackers, mailing lists, etc.
Similarly, CNCF has member fees and other income streams, which are used to run the foundation and provide resources for the projects. These days, most software is not built on laptops; it is run and tested on hundreds of machines on the cloud, and that requires money. Creating marketing campaigns, brand designs, distributing stickers, etc. takes money, and some foundations can assist with that as well. At its core, foundations implement the right processes to interact with users, developers, and control mechanisms and ensure distribution of available financial resources to open source projects for the common good.
If users of open source projects can donate money and the foundations can distribute it in a fair way, what is missing?
What is missing is a direct, transparent, trusted, decentralized, automated bidirectional link for transfer of value between the open source producers and the open source consumer. Currently, the link is either unidirectional or indirect:
* **Unidirectional** : A developer (think of a "developer" as any role that is involved in the production, maintenance, and distribution of software) can use their brain juice and devote time to do a contribution and share that value with all open source users. But there is no reverse link.
* **Indirect** : If there is a bug that affects a specific user/company, the options are:
* To have in-house developers to fix the bug and do a pull request. That is ideal, but it not always possible to hire in-house developers who are knowledgeable about hundreds of open source projects used daily.
* To hire a freelancer specializing in that specific open source project and pay for the services. Ideally, the freelancer is also a committer for the open source project and can directly change the project code quickly. Otherwise, the fix might not ever make it to the project.
* To approach a company providing services around the open source project. Such companies typically employ open source committers to influence and gain credibility in the community and offer products, expertise, and professional services.
The third option has been a successful [model][4] for sustaining many open source projects. Whether they provide services (training, consulting, workshops), support, packaging, open core, or SaaS, there are companies that employ hundreds of staff members who work on open source full time. There is a long [list of companies][5] that have managed to build a successful open source business model over the years, and that list is growing steadily.
The companies that back open source projects play an important role in the ecosystem: They are the catalyst between the open source projects and the users. The ones that add real value do more than just package software nicely; they can identify user needs and technology trends, and they create a full stack and even an ecosystem of open source projects to address these needs. They can take a boring project and support it for years. If there is a missing piece in the stack, they can start an open source project from scratch and build a community around it. They can acquire a closed source software company and open source the projects (here I got a little carried away, but yes, I'm talking about my employer, [Red Hat][6]).
To summarize, with the commercial open source model, projects are officially or unofficially managed and controlled by a very few individuals or companies that monetize them and give back to the ecosystem by ensuring the project is successful. It is a win-win-win for open source developers, managing companies, and end users. The alternative is inactive projects and expensive closed source software.
### Self-sustaining, decentralized open source
For a project to become part of a reputable foundation, it must conform to certain criteria. For example, ASF and CNCF require incubation and graduation processes, respectively, where apart from all the technical and formal requirements, a project must have a healthy number of active committer and users. And that is the essence of forming a sustainable open source project. Having source code on GitHub is not the same thing as having an active open source project. The latter requires committers who write the code and users who use the code, with both groups enforcing each other continuously by exchanging value and forming an ecosystem where everybody benefits. Some project ecosystems might be tiny and short-lived, and some may consist of multiple projects and competing service providers, with very complex interactions lasting for many years. But as long as there is an exchange of value and everybody benefits from it, the project is developed, maintained, and sustained.
If you look at ASF [Attic][7], you will find projects that have reached their end of life. When a project is no longer technologically fit for its purpose, it is usually its natural end. Similarly, in the ASF [Incubator][8], you will find tons of projects that never graduated but were instead retired. Typically, these projects were not able to build a large enough community because they are too specialized or there are better alternatives available.
But there are also cases where projects with high potential and superior technology cannot sustain themselves because they cannot form or maintain a functioning ecosystem for the exchange of value. The open source model and the foundations do not provide a framework and mechanisms for developers to get paid for their work or for users to get their requests heard. There isn’t a common value commitment framework for either party. As a result, some projects can sustain themselves only in the context of commercial open source, where a company acts as an intermediary and value adder between developers and users. That adds another constraint and requires a service provider company to sustain some open source projects. Ideally, users should be able to express their interest in a project and developers should be able to show their commitment to the project in a transparent and measurable way, which forms a community with common interest and intent for the exchange of value.
Imagine there is a model with mechanisms and tools that enable direct interaction between open source users and developers. This includes not only code contributions through pull requests, questions over the mailing lists, GitHub stars, and stickers on laptops, but also other ways that allow users to influence projects' destinies in a richer, more self-controlled and transparent manner.
This model could include incentives for actions such as:
* Funding open source projects directly rather than through software foundations
* Influencing the direction of projects through voting (by token holders)
* Feature requests driven by user needs
* On-time pull request merges
* Bounties for bug hunts
* Better test coverage incentives
* Up-to-date documentation rewards
* Long-term support guarantees
* Timely security fixes
* Expert assistance, support, and services
* Budget for evangelism and promotion of the projects
* Budget for regular boring activities
* Fast email and chat assistance
* Full visibility of the overall project findings, etc.
If you haven't guessed, I'm talking about using blockchain and [smart contracts][9] to allow such interactions between users and developers—smart contracts that will give power to the hand of token holders to influence projects.
![blockchain_in_open_source_ecosystem.png][11]
The usage of blockchain in the open source ecosystem
Existing channels in the open source ecosystem provide ways for users to influence projects through financial commitments to service providers or other limited means through the foundations. But the addition of blockchain-based technology to the open source ecosystem could open new channels for interaction between users and developers. I'm not saying this will replace the commercial open source model; most companies working with open source do many things that cannot be replaced by smart contracts. But smart contracts can spark a new way of bootstrapping new open source projects, giving a second life to commodity projects that are a burden to maintain. They can motivate developers to apply boring pull requests, write documentation, get tests to pass, etc., providing a direct value exchange channel between users and open source developers. Blockchain can add new channels to help open source projects grow and become self-sustaining in the long term, even when company backing is not feasible. It can create a new complementary model for self-sustaining open source projects—a win-win.
### Tokenizing open source
There are already a number of initiatives aiming to tokenize open source. Some focus only on an open source model, and some are more generic but apply to open source development as well:
* [Gitcoin][12] \- grow open source, one of the most promising ones in this area.
* [Oscoin][13] \- cryptocurrency for open source
* [Open collective][14] \- a platform for supporting open source projects.
* [FundYourselfNow][15] \- Kickstarter and ICOs for projects.
* [Kauri][16] \- support for open source project documentation.
* [Liberapay][17] \- a recurrent donations platform.
* [FundRequest][18] \- a decentralized marketplace for open source collaboration.
* [CanYa][19] \- recently acquired [Bountysource][20], now the world’s largest open source P2P bounty platform.
* [OpenGift][21] \- a new model for open source monetization.
* [Hacken][22] \- a white hat token for hackers.
* [Coinlancer][23] \- a decentralized job market.
* [CodeFund][24] \- an open source ad platform.
* [IssueHunt][25] \- a funding platform for open source maintainers and contributors.
* [District0x 1Hive][26] \- a crowdfunding and curation platform.
* [District0x Fixit][27] \- github bug bounties.
This list is varied and growing rapidly. Some of these projects will disappear, others will pivot, but a few will emerge as the [SourceForge][28], the ASF, the GitHub of the future. That doesn't necessarily mean they'll replace these platforms, but they'll complement them with token models and create a richer open source ecosystem. Every project can pick its distribution model (license), governing model (foundation), and incentive model (token). In all cases, this will pump fresh blood to the open source world.
### The future is open and decentralized
* Software is eating the world.
* Every company is a software company.
* Open source is where innovation happens.
Given that, it is clear that open source is too big to fail and too important to be controlled by a few or left to its own destiny. Open source is a shared-resource system that has value to all, and more importantly, it must be managed as such. It is only a matter of time until every company on earth will want to have a stake and a say in the open source world. Unfortunately, we don't have the tools and the habits to do it yet. Such tools would allow anybody to show their appreciation or ignorance of software projects. It would create a direct and faster feedback loop between producers and consumers, between developers and users. It will foster innovation—innovation driven by user needs and expressed through token metrics.
--------------------------------------------------------------------------------
via: https://opensource.com/article/18/8/open-source-tokenomics
作者：[Bilgin lbryam][a]
选题：[lujun9972](https://github.com/lujun9972)
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]:https://opensource.com/users/bibryam
[1]:https://en.wikipedia.org/wiki/Satoshi_Nakamoto
[2]:https://www.apache.org/
[3]:https://www.cncf.io/
[4]:https://medium.com/open-consensus/3-oss-business-model-progressions-dafd5837f2d
[5]:https://docs.google.com/spreadsheets/d/17nKMpi_Dh5slCqzLSFBoWMxNvWiwt2R-t4e_l7LPLhU/edit#gid=0
[6]:http://jobs.redhat.com/
[7]:https://attic.apache.org/
[8]:http://incubator.apache.org/
[9]:https://en.wikipedia.org/wiki/Smart_contract
[10]:/file/404421
[11]:https://opensource.com/sites/default/files/uploads/blockchain_in_open_source_ecosystem.png (blockchain_in_open_source_ecosystem.png)
[12]:https://gitcoin.co/
[13]:http://oscoin.io/
[14]:https://opencollective.com/opensource
[15]:https://www.fundyourselfnow.com/page/about
[16]:https://kauri.io/
[17]:https://liberapay.com/
[18]:https://fundrequest.io/
[19]:https://canya.io/
[20]:https://www.bountysource.com/
[21]:https://opengift.io/pub/
[22]:https://hacken.io/
[23]:https://www.coinlancer.com/home
[24]:https://codefund.io/
[25]:https://issuehunt.io/
[26]:https://blog.district0x.io/district-proposal-spotlight-1hive-283957f57967
[27]:https://github.com/district0x/district-proposals/issues/177
[28]:https://sourceforge.net/

116
sources/talk/20180816 Debian Turns 25- Here are Some Interesting Facts About Debian Linux.md
View File

@ -1,116 +0,0 @@
Debian Turns 25! Here are Some Interesting Facts About Debian Linux
======
One of the oldest Linux distribution still in development, Debian has just turned 25. Let’s have a look at some interesting facts about this awesome FOSS project.
### 10 Interesting facts about Debian Linux
![Interesting facts about Debian Linux][1]
The facts presented here have been collected from various sources available from the internet. They are true to my knowledge, but in case of any error, please remind me to update the article.
#### 1\. One of the oldest Linux distributions still under active development
[Debian project][2] was announced on 16th August 1993 by Ian Murdock, Debian Founder. Like Linux creator [Linus Torvalds][3], Ian was a college student when he announced Debian project.
![](https://farm6.staticflickr.com/5710/20006308374_7f51ae2a5c_z.jpg)
#### 2\. Some people get tattoo while some name their project after their girlfriend’s name
The project was named by combining the name of Ian and his then-girlfriend Debra Lynn. Ian and Debra got married and had three children. Debra and Ian got divorced in 2008.
#### 3\. Ian Murdock: The Maverick behind the creation of Debian project
![Debian Founder Ian Murdock][4]
Ian Murdock
[Ian Murdock][5] led the Debian project from August 1993 until March 1996. He shaped Debian into a community project based on the principals of Free Software. The [Debian Manifesto][6] and the [Debian Social Contract][7] are still governing the project.
He founded a commercial Linux company called [Progeny Linux Systems][8] and worked for a number of Linux related companies such as Sun Microsystems, Linux Foundation and Docker.
Sadly, [Ian committed suicide in December 2015][9]. His contribution to Debian is certainly invaluable.
#### 4\. Debian is a community project in the true sense
Debian is a community based project in true sense. No one ‘owns’ Debian. Debian is being developed by volunteers from all over the world. It is not a commercial project, backed by corporates like many other Linux distributions.
Debian Linux distribution is composed of Free Software only. It’s one of the few Linux distributions that is true to the spirit of [Free Software][10] and takes proud in being called a GNU/Linux distribution.
Debian has its non-profit organization called [Software in Public Interest][11] (SPI). Along with Debian, SPI supports many other open source projects financially.
#### 5\. Debian and its 3 branches
Debian has three branches or versions: Debian Stable, Debian Unstable (Sid) and Debian Testing.
Debian Stable, as the name suggests, is the stable branch that has all the software and packages well tested to give you a rock solid stable system. Since it takes time before a well-tested software lands in the stable branch, Debian Stable often contains older versions of programs and hence people joke that Debian Stable means stale.
[Debian Unstable][12] codenamed Sid is the version where all the development of Debian takes place. This is where the new packages first land or developed. After that, these changes are propagated to the testing version.
[Debian Testing][13] is the next release after the current stable release. If the current stable release is N, Debian testing would be the N+1 release. The packages from Debian Unstable are tested in this version. After all the new changes are well tested, Debian Testing is then ‘promoted’ as the new Stable version.
There is no strict release schedule for Debian.
#### 7\. There was no Debian 1.0 release
Debian 1.0 was never released. The CD vendor, InfoMagic, accidentally shipped a development release of Debian and entitled it 1.0 in 1996. To prevent confusion between the CD version and the actual Debian release, the Debian Project renamed its next release to “Debian 1.1”.
#### 8\. Debian releases are codenamed after Toy Story characters
![Toy Story Characters][14]
Debian releases are codenamed after the characters from Pixar’s hit animation movie series [Toy Story][15].
Debian 1.1 was the first release with a codename. It was named Buzz after the Toy Story character Buzz Lightyear.
It was in 1996 and [Bruce Perens][16] had taken over leadership of the Project from Ian Murdock. Bruce was working at Pixar at the time.
This trend continued and all the subsequent releases had codenamed after Toy Story characters. For example, the current stable release is Stretch while the upcoming release has been codenamed Buster.
The unstable Debian version is codenamed Sid. This character in Toy Story is a kid with emotional problems and he enjoys breaking toys. This is symbolic in the sense that Debian Unstable might break your system with untested packages.
#### 9\. Debian also has a BSD ditribution
Debian is not limited to Linux. Debian also has a distribution based on FreeBSD kernel. It is called [Debian GNU/kFreeBSD][17].
#### 10\. Google uses Debian
[Google uses Debian][18] as its in-house development platform. Earlier, Google used a customized version of Ubuntu as its development platform. Recently they opted for Debian based gLinux.
#### Happy 25th birthday Debian
![Happy 25th birthday Debian][19]
I hope you liked these little facts about Debian. Stuff like these are reasons why people love Debian.
I wish a very happy 25th birthday to Debian. Please continue to be awesome. Cheers :)
--------------------------------------------------------------------------------
via: https://itsfoss.com/debian-facts/
作者：[Abhishek Prakash][a]
选题：[lujun9972](https://github.com/lujun9972)
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://itsfoss.com/author/abhishek/
[1]:https://4bds6hergc-flywheel.netdna-ssl.com/wp-content/uploads/2018/08/Interesting-facts-about-debian.jpeg
[2]:https://www.debian.org
[3]:https://itsfoss.com/linus-torvalds-facts/
[4]:https://4bds6hergc-flywheel.netdna-ssl.com/wp-content/uploads/2018/08/ian-murdock.jpg
[5]:https://en.wikipedia.org/wiki/Ian_Murdock
[6]:https://www.debian.org/doc/manuals/project-history/ap-manifesto.en.html
[7]:https://www.debian.org/social_contract
[8]:https://en.wikipedia.org/wiki/Progeny_Linux_Systems
[9]:https://itsfoss.com/ian-murdock-dies-mysteriously/
[10]:https://www.fsf.org/
[11]:https://www.spi-inc.org/
[12]:https://www.debian.org/releases/sid/
[13]:https://www.debian.org/releases/testing/
[14]:https://4bds6hergc-flywheel.netdna-ssl.com/wp-content/uploads/2018/08/toy-story-characters.jpeg
[15]:https://en.wikipedia.org/wiki/Toy_Story_(franchise)
[16]:https://perens.com/about-bruce-perens/
[17]:https://wiki.debian.org/Debian_GNU/kFreeBSD
[18]:https://itsfoss.com/goobuntu-glinux-google/
[19]:https://4bds6hergc-flywheel.netdna-ssl.com/wp-content/uploads/2018/08/happy-25th-birthday-Debian.jpeg

97
sources/talk/20181004 Interview With Peter Ganten, CEO of Univention GmbH.md
View File

@ -1,97 +0,0 @@
Interview With Peter Ganten, CEO of Univention GmbH
======
I have been asking the Univention team to share the behind-the-scenes story of [**Univention**][1] for a couple of months. Finally, today we got the interview of **Mr. Peter H. Ganten** , CEO of Univention GmbH. Despite his busy schedule, in this interview, he shares what he thinks of the Univention project and its impact on open source ecosystem, what open source developers and companies will need to do to keep thriving and what are the biggest challenges for open source projects.
**OSTechNix: What’s your background and why have you founded Univention?**
**Peter Ganten:** I studied physics and psychology. In psychology I was a research assistant and coded evaluation software. I realized how important it is that results have to be disclosed in order to verify or falsify them. The same goes for the code that leads to the results. This brought me into contact with Open Source Software (OSS) and Linux.
![](https://www.ostechnix.com/wp-content/uploads/2018/10/peter-ganten-interview.jpg)
I was a kind of technical lab manager and I had the opportunity to try out a lot, which led to my book about Debian. That was still in the New Economy era where the first business models emerged on how to make money with Open Source. When the bubble burst, I had the plan to make OSS a solid business model without venture capital but with Hanseatic business style – seriously, steadily, no bling bling.
**What were the biggest challenges at the beginning?**
When I came from the university, the biggest challenge clearly was to gain entrepreneurial and business management knowledge. I quickly learned that it’s not about Open Source software as an end to itself but always about customer value, and the benefits OSS offers its customers. We all had to learn a lot.
In the beginning, we expected that Linux on the desktop would become established in a similar way as Linux on the server. However, this has not yet been proven true. The replacement has happened with Android and the iPhone. Our conclusion then was to change our offerings towards ID management and enterprise servers.
**Why does UCS matter? And for whom makes it sense to use it?**
There is cool OSS in all areas, but many organizations are not capable to combine it all together and make it manageable. For the basic infrastructure (Windows desktops, users, user rights, roles, ID management, apps) we need a central instance to which groupware, CRM etc. is connected. Without Univention this would have to be laboriously assembled and maintained manually. This is possible for very large companies, but far too complex for many other organizations.
[**UCS**][2] can be used out of the box and is scalable. That’s why it’s becoming more and more popular – more than 10,000 organizations are using UCS already today.
**Who are your users and most important clients? What do they love most about UCS?**
The Core Edition is free of charge and used by organizations from all sectors and industries such as associations, micro-enterprises, universities or large organizations with thousands of users. In the enterprise environment, where Long Term Servicing (LTS) and professional support are particularly important, we have organizations ranging in size from 30-50 users to several thousand users. One of the target groups is the education system in Germany. In many large cities and within their school administrations UCS is used, for example, in Cologne, Hannover, Bremen, Kassel and in several federal states. They are looking for manageable IT and apps for schools. That’s what we offer, because we can guarantee these authorities full control over their users’ identities.
Also, more and more cloud service providers and MSPs want to take UCS to deliver a selection of cloud-based app solutions.
**Is UCS 100% Open Source? If so, how can you run a profitable business selling it?**
Yes, UCS is 100% Open Source, every line, the whole code is OSS. You can download and use UCS Core Edition for **FREE!**
We know that in large, complex organizations, vendor support and liability is needed for LTS, SLAs, and we offer that with our Enterprise subscriptions and consulting services. We don’t offer these in the Core Edition.
**And what are you giving back to the OS community?**
A lot. We are involved in the Debian team and co-finance the LTS maintenance for Debian. For important OS components in UCS like [**OpenLDAP**][3], Samba or KVM we co-finance the development or have co-developed them ourselves. We make it all freely available.
We are also involved on the political level in ensuring that OSS is used. We are engaged, for example, in the [**Free Software Foundation Europe (FSFE)**][4] and the [**German Open Source Business Alliance**][5], of which I am the chairman. We are working hard to make OSS more successful.
**How can I get started with UCS?**
It’s easy to get started with the Core Edition, which, like the Enterprise Edition, has an App Center and can be easily installed on your own hardware or as an appliance in a virtual machine. Just [**download Univention ISO**][6] and install it as described in the below link.
Alternatively, you can try the [**UCS Online Demo**][7] to get a first impression of Univention Corporate Server without actually installing it on your system.
**What do you think are the biggest challenges for Open Source?**
There is a certain attitude you can see over and over again even in bigger projects: OSS alone is viewed as an almost mandatory prerequisite for a good, sustainable, secure and trustworthy IT solution – but just having decided to use OSS is no guarantee for success. You have to carry out projects professionally and cooperate with the manufacturers. A danger is that in complex projects people think: “Oh, OSS is free, I just put it all together by myself”. But normally you do not have the know-how to successfully implement complex software solutions. You would never proceed like this with Closed Source. There people think: “Oh, the software costs 3 $ millions, so it’s okay if I have to spend another 300,000 Dollars on consultants.”
At OSS this is different. If such projects fail and leave burnt ground behind, we have to explain again and again that the failure of such projects is not due to the nature of OSS but to its poor implementation and organization in a specific project: You have to conclude reasonable contracts and involve partners as in the proprietary world, but you’ll gain a better solution.
Another challenge: We must stay innovative, move forward, attract new people who are enthusiastic about working on projects. That’s sometimes a challenge. For example, there are a number of proprietary cloud services that are good but lead to extremely high dependency. There are approaches to alternatives in OSS, but no suitable business models yet. So it’s hard to find and fund developers. For example, I can think of Evernote and OneNote for which there is no reasonable OSS alternative.
**And what will the future bring for Univention?**
I don’t have a crystal ball, but we are extremely optimistic. We see a very high growth potential in the education market. More OSS is being made in the public sector, because we have repeatedly experienced the dead ends that can be reached if we solely rely on Closed Source.
Overall, we will continue our organic growth at double-digit rates year after year.
UCS and its core functionalities of identity management, infrastructure management and app center will increasingly be offered and used from the cloud as a managed service. We will support our technology in this direction, e.g., through containers, so that a hypervisor or bare metal is not always necessary for operation.
**You have been the CEO of Univention for a long time. What keeps you motivated?**
I have been the CEO of Univention for more than 16 years now. My biggest motivation is to realize that something is moving. That we offer the better way for IT. That the people who go this way with us are excited to work with us. I go home satisfied in the evening (of course not every evening). It’s totally cool to work with the team I have. It motivates and pushes you every time I need it myself.
I’m a techie and nerd at heart, I enjoy dealing with technology. So I’m totally happy at this place and I’m grateful to the world that I can do whatever I want every day. Not everyone can say that.
**Who gives you inspiration?**
My employees, the customers and the Open Source projects. The exchange with other people.
The motivation behind everything is that we want to make sure that mankind will be able to influence and change the IT that surrounds us today and in the future just the way we want it and we thinks it’s good. We want to make a contribution to this. That is why Univention is there. That is important to us every day.
--------------------------------------------------------------------------------
via: https://www.ostechnix.com/interview-with-peter-ganten-ceo-of-univention-gmbh/
作者：[SK][a]
选题：[lujun9972](https://github.com/lujun9972)
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.ostechnix.com/author/sk/
[1]: https://www.ostechnix.com/introduction-univention-corporate-server/
[2]: https://www.univention.com/products/ucs/
[3]: https://www.ostechnix.com/redhat-and-suse-announced-to-withdraw-support-for-openldap/
[4]: https://fsfe.org/
[5]: https://osb-alliance.de/
[6]: https://www.univention.com/downloads/download-ucs/
[7]: https://www.univention.com/downloads/ucs-online-demo/

76
sources/talk/20190228 Why CLAs aren-t good for open source.md
View File

@ -1,76 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Why CLAs aren't good for open source)
[#]: via: (https://opensource.com/article/19/2/cla-problems)
[#]: author: (Richard Fontana https://opensource.com/users/fontana)
Why CLAs aren't good for open source
======
Few legal topics in open source are as controversial as contributor license agreements.
![](https://opensource.com/sites/default/files/styles/image-full-size/public/lead-images/write-hand_0.jpg?itok=Uw5RJD03)
Few legal topics in open source are as controversial as [contributor license agreements][1] (CLAs). Unless you count the special historical case of the [Fedora Project Contributor Agreement][2] (which I've always seen as an un-CLA), or, like [Karl Fogel][3], you classify the [DCO][4] as a [type of CLA][5], today Red Hat makes no use of CLAs for the projects it maintains.
It wasn't always so. Red Hat's earliest projects followed the traditional practice I've called "inbound=outbound," in which contributions to a project are simply provided under the project's open source license with no execution of an external, non-FOSS contract required. But in the early 2000s, Red Hat began experimenting with the use of contributor agreements. Fedora started requiring contributors to sign a CLA based on the widely adapted [Apache ICLA][6], while a Free Software Foundation-derived copyright assignment agreement and a pair of bespoke CLAs were inherited from the Cygnus and JBoss acquisitions, respectively. We even took [a few steps][7] towards adopting an Apache-style CLA across the rapidly growing set of Red Hat-led projects.
This came to an end, in large part because those of us on the Red Hat legal team heard and understood the concerns and objections raised by Red Hat engineers and the wider technical community. We went on to become de facto leaders of what some have called the anti-CLA movement, marked notably by our [opposition to Project Harmony][8] and our [efforts][9] to get OpenStack to replace its CLA with the DCO. (We [reluctantly][10] sign tolerable upstream project CLAs out of practical necessity.)
### Why CLAs are problematic
Our choice not to use CLAs is a reflection of our values as an authentic open source company with deep roots in the free software movement. Over the years, many in the open source community have explained why CLAs, and the very similar mechanism of copyright assignment, are a bad policy for open source.
One reason is the red tape problem. Normally, open source development is characterized by frictionless contribution, which is enabled by inbound=outbound without imposition of further legal ceremony or process. This makes it relatively easy for new contributors to get involved in a project, allowing more effective growth of contributor communities and driving technical innovation upstream. Frictionless contribution is a key part of the advantage open source development holds over proprietary alternatives. But frictionless contribution is negated by CLAs. Having to sign an unusual legal agreement before a contribution can be accepted creates a bureaucratic hurdle that slows down development and discourages participation. This cost persists despite the growing use of automation by CLA-using projects.
CLAs also give rise to an asymmetry of legal power among a project's participants, which also discourages the growth of strong contributor and user communities around a project. With Apache-style CLAs, the company or organization leading the project gets special rights that other contributors do not receive, while those other contributors must shoulder certain legal obligations (in addition to the red tape burden) from which the project leader is exempt. The problem of asymmetry is most severe in copyleft projects, but it is present even when the outbound license is permissive.
When assessing the arguments for and against CLAs, bear in mind that today, as in the past, the vast majority of the open source code in any product originates in projects that follow the inbound=outbound practice. The use of CLAs by a relatively small number of projects causes collateral harm to all the others by signaling that, for some reason, open source licensing is insufficient to handle contributions flowing into a project.
### The case for CLAs
Since CLAs continue to be a minority practice and originate from outside open source community culture, I believe that CLA proponents should bear the burden of explaining why they are necessary or beneficial relative to their costs. I suspect that most companies using CLAs are merely emulating peer company behavior without critical examination. CLAs have an understandable, if superficial, appeal to risk-averse lawyers who are predisposed to favor greater formality, paper, and process regardless of the business costs. Still, some arguments in favor of CLAs are often advanced and deserve consideration.
**Easy relicensing:** If administered appropriately, Apache-style CLAs give the project steward effectively unlimited power to sublicense contributions under terms of the steward's choice. This is sometimes seen as desirable because of the potential need to relicense a project under some other open source license. But the value of easy relicensing has been greatly exaggerated by pointing to a few historical cases involving major relicensing campaigns undertaken by projects with an unusually large number of past contributors (all of which were successful without the use of a CLA). There are benefits in relicensing being hard because it results in stable legal expectations around a project and encourages projects to consult their contributor communities before undertaking significant legal policy changes. In any case, most inbound=outbound open source projects never attempt to relicense during their lifetime, and for the small number that do, relicensing will be relatively painless because typically the number of past contributors to contact will not be large.
**Provenance tracking:** It is sometimes claimed that CLAs enable a project to rigorously track the provenance of contributions, which purportedly has some legal benefit. It is unclear what is achieved by the use of CLAs in this regard that is not better handled through such non-CLA means as preserving Git commit history. And the DCO would seem to be much better suited to tracking contributions, given that it is normally used on a per-commit basis, while CLAs are signed once per contributor and are administratively separate from code contributions. Moreover, provenance tracking is often described as though it were a benefit for the public, yet I know of no case where a project provides transparent, ready public access to CLA acceptance records.
**License revocation:** Some CLA advocates warn of the prospect that a contributor may someday attempt to revoke a past license grant. To the extent that the concern is about largely judgment-proof individual contributors with no corporate affiliation, it is not clear why an Apache-style CLA provides more meaningful protection against this outcome compared to the use of an open source license. And, as with so many of the legal risks raised in discussions of open source legal policy, this appears to be a phantom risk. I have heard of only a few purported attempts at license revocation over the years, all of which were resolved quickly when the contributor backed down in the face of community pressure.
**Unauthorized employee contribution:** This is a special case of the license revocation issue and has recently become a point commonly raised by CLA advocates. When an employee contributes to an upstream project, normally the employer owns the copyrights and patents for which the project needs licenses, and only certain executives are authorized to grant such licenses. Suppose an employee contributed proprietary code to a project without approval from the employer, and the employer later discovers this and demands removal of the contribution or sues the project's users. This risk of unauthorized contributions is thought to be minimized by use of something like the [Apache CCLA][11] with its representations and signature requirement, coupled with some adequate review process to ascertain that the CCLA signer likely was authorized to sign (a step which I suspect is not meaningfully undertaken by most CLA-using companies).
Based on common sense and common experience, I contend that in nearly all cases today, employee contributions are done with the actual or constructive knowledge and consent of the employer. If there were an atmosphere of high litigation risk surrounding open source software, perhaps this risk should be taken more seriously, but litigation arising out of open source projects remains remarkably uncommon.
More to the point, I know of no case where an allegation of copyright or patent infringement against an inbound=outbound project, not stemming from an alleged open source license violation, would have been prevented by use of a CLA. Patent risk, in particular, is often cited by CLA proponents when pointing to the risk of unauthorized contributions, but the patent license grants in Apache-style CLAs are, by design, quite narrow in scope. Moreover, corporate contributions to an open source project will typically be few in number, small in size (and thus easily replaceable), and likely to be discarded as time goes on.
### Alternatives
If your company does not buy into the anti-CLA case and cannot get comfortable with the simple use of inbound=outbound, there are alternatives to resorting to an asymmetric and administratively burdensome Apache-style CLA requirement. The use of the DCO as a complement to inbound=outbound addresses at least some of the concerns of risk-averse CLA advocates. If you must use a true CLA, there is no need to use the Apache model (let alone a [monstrous derivative][10] of it). Consider the non-specification core of the [Eclipse Contributor Agreement][12]—essentially the DCO wrapped inside a CLA—or the Software Freedom Conservancy's [Selenium CLA][13], which merely ceremonializes an inbound=outbound contribution policy.
--------------------------------------------------------------------------------
via: https://opensource.com/article/19/2/cla-problems
作者：[Richard Fontana][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://opensource.com/users/fontana
[b]: https://github.com/lujun9972
[1]: https://opensource.com/article/18/3/cla-vs-dco-whats-difference
[2]: https://opensource.com/law/10/6/new-contributor-agreement-fedora
[3]: https://www.red-bean.com/kfogel/
[4]: https://developercertificate.org
[5]: https://producingoss.com/en/contributor-agreements.html#developer-certificate-of-origin
[6]: https://www.apache.org/licenses/icla.pdf
[7]: https://www.freeipa.org/page/Why_CLA%3F
[8]: https://opensource.com/law/11/7/trouble-harmony-part-1
[9]: https://wiki.openstack.org/wiki/OpenStackAndItsCLA
[10]: https://opensource.com/article/19/1/cla-proliferation
[11]: https://www.apache.org/licenses/cla-corporate.txt
[12]: https://www.eclipse.org/legal/ECA.php
[13]: https://docs.google.com/forms/d/e/1FAIpQLSd2FsN12NzjCs450ZmJzkJNulmRC8r8l8NYwVW5KWNX7XDiUw/viewform?hl=en_US&formkey=dFFjXzBzM1VwekFlOWFWMjFFRjJMRFE6MQ#gid=0

45
sources/talk/20190311 Discuss everything Fedora.md
View File

@ -1,45 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Discuss everything Fedora)
[#]: via: (https://fedoramagazine.org/discuss-everything-fedora/)
[#]: author: (Ryan Lerch https://fedoramagazine.org/introducing-flatpak/)
Discuss everything Fedora
======
![](https://fedoramagazine.org/wp-content/uploads/2019/03/fedora-discussion-816x345.jpg)
Are you interested in how Fedora is being developed? Do you want to get involved, or see what goes into making a release? You want to check out [Fedora Discussion][1]. It is a relatively new place where members of the Fedora Community meet to discuss, ask questions, and interact. Keep reading for more information.
Note that the Fedora Discussion system is mainly aimed at contributors. If you have questions on using Fedora, check out [Ask Fedora][2] (which is being migrated in the future).
![][3]
Fedora Discussion is a forum and discussion site that uses the [Discourse open source discussion platform][4].
There are already several categories useful for Fedora users, including [Desktop][5] (covering Fedora Workstation, Fedora Silverblue, KDE, XFCE, and more) and the [Server, Cloud, and IoT][6] category . Additionally, some of the [Fedora Special Interest Groups (SIGs) have discussions as well][7]. Finally, the [Fedora Friends][8] category helps you connect with other Fedora users and Community members by providing discussions about upcoming meetups and hackfests.
--------------------------------------------------------------------------------
via: https://fedoramagazine.org/discuss-everything-fedora/
作者：[Ryan Lerch][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://fedoramagazine.org/introducing-flatpak/
[b]: https://github.com/lujun9972
[1]: https://discussion.fedoraproject.org/
[2]: https://ask.fedoraproject.org
[3]: https://fedoramagazine.org/wp-content/uploads/2019/03/discussion-screenshot-1024x663.png
[4]: https://www.discourse.org/about
[5]: https://discussion.fedoraproject.org/c/desktop
[6]: https://discussion.fedoraproject.org/c/server
[7]: https://discussion.fedoraproject.org/c/sigs
[8]: https://discussion.fedoraproject.org/c/friends

143
sources/talk/20190322 How to save time with TiDB.md
View File

@ -1,143 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (How to save time with TiDB)
[#]: via: (https://opensource.com/article/19/3/how-save-time-tidb)
[#]: author: (Morgan Tocker https://opensource.com/users/morgo)
How to save time with TiDB
======
TiDB, an open source-compatible, cloud-based database engine, simplifies many of MySQL database administrators' common tasks.
![Team checklist][1]
Last November, I wrote about key [differences between MySQL and TiDB][2], an open source-compatible, cloud-based database engine, from the perspective of scaling both solutions in the cloud. In this follow-up article, I'll dive deeper into the ways [TiDB][3] streamlines and simplifies administration.
If you come from a MySQL background, you may be used to doing a lot of manual tasks that are either not required or much simpler with TiDB.
The inspiration for TiDB came from the founders managing sharded MySQL at scale at some of China's largest internet companies. Since requirements for operating a large system at scale are a key concern, I'll look at some typical MySQL database administrator (DBA) tasks and how they translate to TiDB.
[![TiDB architecture][4]][5]
In [TiDB's architecture][5]:
* SQL processing is separated from data storage. The SQL processing (TiDB) and storage (TiKV) components independently scale horizontally.
* PD (Placement Driver) acts as the cluster manager and stores metadata.
* All components natively provide high availability, with PD and TiKV using the [Raft consensus algorithm][6].
* You can access your data via either MySQL (TiDB) or Spark (TiSpark) protocols.
### Adding/fixing replication slaves
**tl;dr:** It doesn't happen in the same way as in MySQL.
Replication and redundancy of data are automatically managed by TiKV. You also don't need to worry about creating initial backups to seed replicas, as _both_ the provisioning and replication are handled for you.
Replication is also quorum-based using the Raft consensus algorithm, so you don't have to worry about the inconsistency problems surrounding failures that you do with asynchronous replication (the default in MySQL and what many users are using).
TiDB does support its own binary log, so it can be used for asynchronous replication between clusters.
### Optimizing slow queries
**tl;dr:** Still happens in TiDB
There is no real way out of optimizing slow queries that have been introduced by development teams.
As a mitigating factor though, if you need to add breathing room to your database's capacity while you work on optimization, the TiDB's architecture allows you to horizontally scale.
### Upgrades and maintenance
**tl;dr:** Still required, but generally easier
Because the TiDB server is stateless, you can roll through an upgrade and deploy new TiDB servers. Then you can remove the older TiDB servers from the load balancer pool, shutting down them once connections have drained.
Upgrading PD is also quite straightforward since only the PD leader actively answers requests at a time. You can perform a rolling upgrade and upgrade PD's non-leader peers one at a time, and then change the leader before upgrading the final PD server.
For TiKV, the upgrade is marginally more complex. If you want to remove a node, I recommend first setting it to be a follower on each of the regions where it is currently a leader. After that, you can bring down the node without impacting your application. If the downtime is brief, TiKV will recover with its regional peers from the Raft log. In a longer downtime, it will need to re-copy data. This can all be managed for you, though, if you choose to deploy using Ansible or Kubernetes.
### Manual sharding
**tl;dr:** Not required
Manual sharding is mainly a pain on the part of the application developers, but as a DBA, you might have to get involved if the sharding is naive or has problems such as hotspots (many workloads do) that require re-balancing.
In TiDB, re-sharding or re-balancing happens automatically in the background. The PD server observes when data regions (TiKV's term for chunks of data in key-value form) get too small, too big, or too frequently accessed.
You can also explicitly configure PD to store regions on certain TiKV servers. This works really well when combined with MySQL partitioning.
### Capacity planning
**tl;dr:** Much easier
Capacity planning on a MySQL database can be a little bit hard because you need to plan your physical infrastructure requirements two to three years from now. As data grows (and the working set changes), this can be a difficult task. I wouldn't say it completely goes away in the cloud either, since changing a master server's hardware is always hard.
TiDB splits data into approximately 100MiB chunks that it distributes among TiKV servers. Because this increment is much smaller than a full server, it's much easier to move around and redistribute data. It's also possible to add new servers in smaller increments, which is easier on planning.
### Scaling
**tl;dr:** Much easier
This is related to capacity planning and sharding. When we talk about scaling, many people think about very large _systems,_ but that is not exclusively how I think of the problem:
* Scaling is being able to start with something very small, without having to make huge investments upfront on the chance it could become very large.
* Scaling is also a people problem. If a system requires too much internal knowledge to operate, it can become hard to grow as an engineering organization. The barrier to entry for new hires can become very high.
Thus, by providing automatic sharding, TiDB can scale much easier.
### Schema changes (DDL)
**tl;dr:** Mostly better
The data definition language (DDL) supported in TiDB is all online, which means it doesn't block other reads or writes to the system. It also doesn't block the replication stream.
That's the good news, but there are a couple of limitations to be aware of:
* TiDB does not currently support all DDL operations, such as changing the primary key or some "change data type" operations.
* TiDB does not currently allow you to chain multiple DDL changes in the same command, e.g., _ALTER TABLE t1 ADD INDEX (x), ADD INDEX (y)_. You will need to break these queries up into individual DDL queries.
This is an area that we're looking to improve in [TiDB 3.0][7].
### Creating one-off data dumps for the reporting team
**tl;dr:** May not be required
DBAs loathe manual tasks that create one-off exports of data to be consumed by another team, perhaps in an analytics tool or data warehouse.
This is often required when the types of queries that are be executed on the dataset are analytical. TiDB has hybrid transactional/analytical processing (HTAP) capabilities, so in many cases, these queries should work fine. If your analytics team is using Spark, you can also use the [TiSpark][8] connector to allow them to connect directly to TiKV.
This is another area we are improving with [TiFlash][7], a column store accelerator. We are also working on a plugin system to support external authentication. This will make it easier to manage access by the reporting team.
### Conclusion
In this post, I looked at some common MySQL DBA tasks and how they translate to TiDB. If you would like to learn more, check out our [TiDB Academy course][9] designed for MySQL DBAs (it's free!).
--------------------------------------------------------------------------------
via: https://opensource.com/article/19/3/how-save-time-tidb
作者：[Morgan Tocker][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://opensource.com/users/morgo
[b]: https://github.com/lujun9972
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/lead-images/checklist_todo_clock_time_team.png?itok=1z528Q0y (Team checklist)
[2]: https://opensource.com/article/18/11/key-differences-between-mysql-and-tidb
[3]: https://github.com/pingcap/tidb
[4]: https://opensource.com/sites/default/files/uploads/tidb_architecture.png (TiDB architecture)
[5]: https://pingcap.com/docs/architecture/
[6]: https://raft.github.io/
[7]: https://pingcap.com/blog/tidb-3.0-beta-stability-at-scale/
[8]: https://github.com/pingcap/tispark
[9]: https://pingcap.com/tidb-academy/

60
sources/talk/20190410 Google partners with Intel, HPE and Lenovo for hybrid cloud.md
View File

@ -1,60 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Google partners with Intel, HPE and Lenovo for hybrid cloud)
[#]: via: (https://www.networkworld.com/article/3388062/google-partners-with-intel-hpe-and-lenovo-for-hybrid-cloud.html#tk.rss_all)
[#]: author: (Andy Patrizio https://www.networkworld.com/author/Andy-Patrizio/)
Google partners with Intel, HPE and Lenovo for hybrid cloud
======
Google boosted its on-premises and cloud connections with Kubernetes and serverless computing.
![Ilze Lucero \(CC0\)][1]
Still struggling to get its Google Cloud business out of single-digit marketshare, Google this week introduced new partnerships with Lenovo and Intel to help bolster its hybrid cloud offerings, both built on Google’s Kubernetes container technology.
At Google’s Next ’19 show this week, Intel and Google said they will collaborate on Google's Anthos, a new reference design based on the second-Generation Xeon Scalable processor introduced last week and an optimized Kubernetes software stack designed to deliver increased workload portability between public and private cloud environments.
**[ Read also:[What hybrid cloud means in practice][2] | Get regularly scheduled insights: [Sign up for Network World newsletters][3] ]**
As part the Anthos announcement, Hewlett Packard Enterprise (HPE) said it has validated Anthos on its ProLiant servers, while Lenovo has done the same for its ThinkAgile platform. This solution will enable customers to get a consistent Kubernetes experience between Google Cloud and their on-premises HPE or Lenovo servers. No official word from Dell yet, but they can’t be far behind.
Users will be able to manage their Kubernetes clusters and enforce policy consistently across environments – either in the public cloud or on-premises. In addition, Anthos delivers a fully integrated stack of hardened components, including OS and container runtimes that are tested and validated by Google, so customers can upgrade their clusters with confidence and minimize downtime.
### What is Google Anthos?
Google formally introduced [Anthos][4] at this year’s show. Anthos, formerly Cloud Services Platform, is meant to allow users to run their containerized applications without spending time on building, managing, and operating Kubernetes clusters. It runs both on Google Cloud Platform (GCP) with Google Kubernetes Engine (GKE) and in your data center with GKE On-Prem. Anthos will also let you manage workloads running on third-party clouds such as Amazon Web Services (AWS) and Microsoft Azure.
Google also announced the beta release of Anthos Migrate, which auto-migrates virtual machines (VM) from on-premises or other clouds directly into containers in GKE with minimal effort. This allows enterprises to migrate their infrastructure in one streamlined motion, without upfront modifications to the original VMs or applications.
Intel said it will publish the production design as an Intel Select Solution, as well as a developer platform, making it available to anyone who wants it.
### Serverless environments
Google isn’t stopping with Kubernetes containers, it’s also pushing ahead with serverless environments. [Cloud Run][5] is Google’s implementation of serverless computing, which is something of a misnomer. You still run your apps on servers; you just aren’t using a dedicated physical server. It is stateless, so resources are not allocated until you actually run or use the application.
Cloud Run is a fully serverless offering that takes care of all infrastructure management, including the provisioning, configuring, scaling, and managing of servers. It automatically scales up or down within seconds, even down to zero depending on traffic, ensuring you pay only for the resources you actually use. Cloud Run can be used on GKE, offering the option to run side by side with other workloads deployed in the same cluster.
Join the Network World communities on [Facebook][6] and [LinkedIn][7] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3388062/google-partners-with-intel-hpe-and-lenovo-for-hybrid-cloud.html#tk.rss_all
作者：[Andy Patrizio][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Andy-Patrizio/
[b]: https://github.com/lujun9972
[1]: https://images.idgesg.net/images/article/2018/03/cubes_blocks_squares_containers_ilze_lucero_cc0_via_unsplash_1200x800-100752172-large.jpg
[2]: https://www.networkworld.com/article/3249495/what-hybrid-cloud-mean-practice
[3]: https://www.networkworld.com/newsletters/signup.html
[4]: https://cloud.google.com/blog/topics/hybrid-cloud/new-platform-for-managing-applications-in-todays-multi-cloud-world
[5]: https://cloud.google.com/blog/products/serverless/announcing-cloud-run-the-newest-member-of-our-serverless-compute-stack
[6]: https://www.facebook.com/NetworkWorld/
[7]: https://www.linkedin.com/company/network-world

60
sources/talk/20190410 HPE and Nutanix partner for hyperconverged private cloud systems.md
View File

@ -1,60 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (HPE and Nutanix partner for hyperconverged private cloud systems)
[#]: via: (https://www.networkworld.com/article/3388297/hpe-and-nutanix-partner-for-hyperconverged-private-cloud-systems.html#tk.rss_all)
[#]: author: (Andy Patrizio https://www.networkworld.com/author/Andy-Patrizio/)
HPE and Nutanix partner for hyperconverged private cloud systems
======
Both companies will sell HP ProLiant appliances with Nutanix software but to different markets.
![Hewlett Packard Enterprise][1]
Hewlett Packard Enterprise (HPE) has partnered with Nutanix to offer Nutanix’s hyperconverged infrastructure (HCI) software available as a managed private cloud service and on HPE-branded appliances.
As part of the deal, the two companies will be competing against each other in hardware sales, sort of. If you want the consumption model you get through HPE’s GreenLake, where your usage is metered and you pay for only the time you use it (similar to the cloud), then you would get the ProLiant hardware from HPE.
If you want an appliance model where you buy the hardware outright, like in the traditional sense of server sales, you would get the same ProLiant through Nutanix.
**[ Read also:[What is hybrid cloud computing?][2] and [Multicloud mania: what to know][3] ]**
As it is, HPE GreenLake offers multiple cloud offerings to customers, including virtualization courtesy of VMware and Microsoft. With the Nutanix partnership, HPE is adding Nutanix’s free Acropolis hypervisor to its offerings.
“Customers get to choose an alternative to VMware with this,” said Pradeep Kumar, senior vice president and general manager of HPE’s Pointnext consultancy. “They like the Acropolis license model, since it’s license-free. Then they have choice points so pricing is competitive. Some like VMware, and I think it’s our job to offer them both and they can pick and choose.”
Kumar added that the whole Nutanix stack is 15 to 18% less with Acropolis than a VMware-powered system, since they save on the hypervisor.
The HPE-Nutanix partnership offers a fully managed hybrid cloud infrastructure delivered as a service and deployed in customers’ data centers or co-location facility. The managed private cloud service gives enterprises a hyperconverged environment in-house without having to manage the infrastructure themselves and, more importantly, without the burden of ownership. GreenLake operates more like a lease than ownership.
### HPE GreenLake's private cloud services promise to significantly reduce costs
HPE is pushing hard on GreenLake, which basically mimics cloud platform pricing models of paying for what you use rather than outright ownership. Kumar said HPE projects the consumption model will account for 30% of HPE’s business in the next few years.
GreenLake makes some hefty promises. According to Nutanix-commissioned IDC research, customers will achieve a 60% reduction in the five-year cost of operations, while a HPE-commissioned Forrester report found customers benefit from a 30% Capex savings due to eliminated need for overprovisioning and a 90% reduction in support and professional services costs.
By shifting to an IT as a Service model, HPE claims to provide a 40% increase in productivity by reducing the support load on IT operations staff and to shorten the time to deploy IT projects by 65%.
The two new offerings from the partnership – HPE GreenLake’s private cloud service running Nutanix software and the HPE-branded appliances integrated with Nutanix software – are expected to be available during the 2019 third quarter, the companies said.
Join the Network World communities on [Facebook][4] and [LinkedIn][5] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3388297/hpe-and-nutanix-partner-for-hyperconverged-private-cloud-systems.html#tk.rss_all
作者：[Andy Patrizio][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Andy-Patrizio/
[b]: https://github.com/lujun9972
[1]: https://images.techhive.com/images/article/2015/11/hpe_building-100625424-large.jpg
[2]: https://www.networkworld.com/article/3233132/cloud-computing/what-is-hybrid-cloud-computing.html
[3]: https://www.networkworld.com/article/3252775/hybrid-cloud/multicloud-mania-what-to-know.html
[4]: https://www.facebook.com/NetworkWorld/
[5]: https://www.linkedin.com/company/network-world

76
sources/talk/20190418 Cisco warns WLAN controller, 9000 series router and IOS-XE users to patch urgent security holes.md
View File

@ -1,76 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Cisco warns WLAN controller, 9000 series router and IOS/XE users to patch urgent security holes)
[#]: via: (https://www.networkworld.com/article/3390159/cisco-warns-wlan-controller-9000-series-router-and-iosxe-users-to-patch-urgent-security-holes.html#tk.rss_all)
[#]: author: (Michael Cooney https://www.networkworld.com/author/Michael-Cooney/)
Cisco warns WLAN controller, 9000 series router and IOS/XE users to patch urgent security holes
======
Cisco says unpatched vulnerabilities could lead to DoS attacks, arbitrary code execution, take-over of devices.
![Woolzian / Getty Images][1]
Cisco this week issued 31 security advisories but directed customer attention to “critical” patches for its IOS and IOS XE Software Cluster Management and IOS software for Cisco ASR 9000 Series routers. A number of other vulnerabilities also need attention if customers are running Cisco Wireless LAN Controllers.
The [first critical patch][2] has to do with a vulnerability in the Cisco Cluster Management Protocol (CMP) processing code in Cisco IOS and Cisco IOS XE Software that could allow an unauthenticated, remote attacker to send malformed CMP-specific Telnet options while establishing a Telnet session with an affected Cisco device configured to accept Telnet connections. An exploit could allow an attacker to execute arbitrary code and obtain full control of the device or cause a reload of the affected device, Cisco said.
**[ Also see[What to consider when deploying a next generation firewall][3]. | Get regularly scheduled insights by [signing up for Network World newsletters][4]. ]**
The problem has a Common Vulnerability Scoring System number of 9.8 out of 10.
According to Cisco, the Cluster Management Protocol utilizes Telnet internally as a signaling and command protocol between cluster members. The vulnerability is due to the combination of two factors:
* The failure to restrict the use of CMP-specific Telnet options only to internal, local communications between cluster members and instead accept and process such options over any Telnet connection to an affected device
* The incorrect processing of malformed CMP-specific Telnet options.
Cisco says the vulnerability can be exploited during Telnet session negotiation over either IPv4 or IPv6. This vulnerability can only be exploited through a Telnet session established _to_ the device; sending the malformed options on Telnet sessions _through_ the device will not trigger the vulnerability.
The company says there are no workarounds for this problem, but disabling Telnet as an allowed protocol for incoming connections would eliminate the exploit vector. Cisco recommends disabling Telnet and using SSH instead. Information on how to do both can be found on the [Cisco Guide to Harden Cisco IOS Devices][5]. For patch information [go here][6].
The second critical patch involves a vulnerability in the sysadmin virtual machine (VM) on Cisco’s ASR 9000 carrier class routers running Cisco IOS XR 64-bit Software could let an unauthenticated, remote attacker access internal applications running on the sysadmin VM, Cisco said in the [advisory][7]. This CVSS also has a 9.8 rating.
**[[Prepare to become a Certified Information Security Systems Professional with this comprehensive online course from PluralSight. Now offering a 10-day free trial!][8] ]**
Cisco said the vulnerability is due to incorrect isolation of the secondary management interface from internal sysadmin applications. An attacker could exploit this vulnerability by connecting to one of the listening internal applications. A successful exploit could result in unstable conditions, including both denial of service (DoS) and remote unauthenticated access to the device, Cisco stated.
Cisco has released [free software updates][6] that address the vulnerability described in this advisory.
Lastly, Cisco wrote that [multiple vulnerabilities][9] in the administrative GUI configuration feature of Cisco Wireless LAN Controller (WLC) Software could let an authenticated, remote attacker cause the device to reload unexpectedly during device configuration when the administrator is using this GUI, causing a DoS condition on an affected device. The attacker would need to have valid administrator credentials on the device for this exploit to work, Cisco stated.
“These vulnerabilities are due to incomplete input validation for unexpected configuration options that the attacker could submit while accessing the GUI configuration menus. An attacker could exploit these vulnerabilities by authenticating to the device and submitting crafted user input when using the administrative GUI configuration feature,” Cisco stated.
“These vulnerabilities have a Security Impact Rating (SIR) of High because they could be exploited when the software fix for the Cisco Wireless LAN Controller Cross-Site Request Forgery Vulnerability is not in place,” Cisco stated. “In that case, an unauthenticated attacker who first exploits the cross-site request forgery vulnerability could perform arbitrary commands with the privileges of the administrator user by exploiting the vulnerabilities described in this advisory.”
Cisco has released [software updates][10] that address these vulnerabilities and said that there are no workarounds.
Join the Network World communities on [Facebook][11] and [LinkedIn][12] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3390159/cisco-warns-wlan-controller-9000-series-router-and-iosxe-users-to-patch-urgent-security-holes.html#tk.rss_all
作者：[Michael Cooney][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Michael-Cooney/
[b]: https://github.com/lujun9972
[1]: https://images.idgesg.net/images/article/2019/02/compromised_data_security_breach_vulnerability_by_woolzian_gettyimages-475563052_2400x1600-100788413-large.jpg
[2]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20170317-cmp
[3]: https://www.networkworld.com/article/3236448/lan-wan/what-to-consider-when-deploying-a-next-generation-firewall.html
[4]: https://www.networkworld.com/newsletters/signup.html
[5]: http://www.cisco.com/c/en/us/support/docs/ip/access-lists/13608-21.html
[6]: https://www.cisco.com/c/en/us/about/legal/cloud-and-software/end_user_license_agreement.html
[7]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190417-asr9k-exr
[8]: https://pluralsight.pxf.io/c/321564/424552/7490?u=https%3A%2F%2Fwww.pluralsight.com%2Fpaths%2Fcertified-information-systems-security-professional-cisspr
[9]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190417-wlc-iapp
[10]: https://www.cisco.com/c/en/us/support/web/tsd-cisco-worldwide-contacts.html
[11]: https://www.facebook.com/NetworkWorld/
[12]: https://www.linkedin.com/company/network-world

58
sources/talk/20190418 Fujitsu completes design of exascale supercomputer, promises to productize it.md
View File

@ -1,58 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Fujitsu completes design of exascale supercomputer, promises to productize it)
[#]: via: (https://www.networkworld.com/article/3389748/fujitsu-completes-design-of-exascale-supercomputer-promises-to-productize-it.html#tk.rss_all)
[#]: author: (Andy Patrizio https://www.networkworld.com/author/Andy-Patrizio/)
Fujitsu completes design of exascale supercomputer, promises to productize it
======
Fujitsu hopes to be the first to offer exascale supercomputing using Arm processors.
![Riken Advanced Institute for Computational Science][1]
Fujitsu and Japanese research institute Riken announced the design for the post-K supercomputer, to be launched in 2021, is complete and that they will productize the design for sale later this year.
The K supercomputer was a massive system, built by Fujitsu and housed at the Riken Advanced Institute for Computational Science campus in Kobe, Japan, with more than 80,000 nodes and using Sparc64 VIIIfx processors, a derivative of the Sun Microsystems Sparc processor developed under a license agreement that pre-dated Oracle buying out Sun in 2010.
**[ Also read:[10 of the world's fastest supercomputers][2] ]**
It was ranked as the top supercomputer when it was launched in June 2011 with a computation speed of over 8 petaflops. And in November 2011, K became the first computer to top 10 petaflops. It was eventually surpassed as the world's fastest supercomputer by the IBM’s Sequoia, but even now, eight years later, it’s still in the top 20 of supercomputers in the world.
### What's in the Post-K supercomputer?
The new system, dubbed “Post-K,” will feature an Arm-based processor called A64FX, a high-performance CPU developed by Fujitsu, designed for exascale systems. The chip is based off the Arm8 design, which is popular in smartphones, with 48 cores plus four “assistant” cores and the ability to access up to 32GB of memory per chip.
A64FX is the first CPU to adopt the Scalable Vector Extension (SVE), an instruction set specifically designed for Arm-based supercomputers. Fujitsu claims A64FX will offer a peak double precision (64-bit) floating point operations performance of over 2.7 teraflops per chip. The system will have one CPU per node and 384 nodes per rack. That comes out to one petaflop per rack.
Contrast that with Summit, the top supercomputer in the world built by IBM for the Oak Ridge National Laboratory using IBM Power9 processors and Nvidia GPUs. A Summit rack has a peak computer of 864 teraflops.
Let me put it another way: IBM’s Power processor and Nvidia’s Tesla are about to get pwned by a derivative chip to the one in your iPhone.
**[[Get certified as an Apple Technical Coordinator with this seven-part online course from PluralSight.][3] ]**
Fujitsu will productize the Post-K design and sell it as the successor to the Fujitsu Supercomputer PrimeHPC FX100. The company said it is also considering measures such as developing an entry-level model that will be easy to deploy, or supplying these technologies to other vendors.
Post-K will be installed in the Riken Center for Computational Science (R-CCS), where the K computer is currently located. The system will be one of the first exascale supercomputers in the world, although the U.S. and China are certainly gunning to be first if only for bragging rights.
Join the Network World communities on [Facebook][4] and [LinkedIn][5] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3389748/fujitsu-completes-design-of-exascale-supercomputer-promises-to-productize-it.html#tk.rss_all
作者：[Andy Patrizio][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Andy-Patrizio/
[b]: https://github.com/lujun9972
[1]: https://images.idgesg.net/images/article/2018/06/riken_advanced_institute_for_computational_science_k-computer_supercomputer_1200x800-100762135-large.jpg
[2]: https://www.networkworld.com/article/3236875/embargo-10-of-the-worlds-fastest-supercomputers.html
[3]: https://pluralsight.pxf.io/c/321564/424552/7490?u=https%3A%2F%2Fwww.pluralsight.com%2Fpaths%2Fapple-certified-technical-trainer-10-11
[4]: https://www.facebook.com/NetworkWorld/
[5]: https://www.linkedin.com/company/network-world

61
sources/talk/20190419 Intel follows AMD-s lead (again) into single-socket Xeon servers.md
View File

@ -1,61 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Intel follows AMD’s lead (again) into single-socket Xeon servers)
[#]: via: (https://www.networkworld.com/article/3390201/intel-follows-amds-lead-again-into-single-socket-xeon-servers.html#tk.rss_all)
[#]: author: (Andy Patrizio https://www.networkworld.com/author/Andy-Patrizio/)
Intel follows AMD’s lead (again) into single-socket Xeon servers
======
Intel's new U series of processors are aimed at the low-end market where one processor is good enough.
![Intel][1]
I’m really starting to wonder who the leader in x86 really is these days because it seems Intel is borrowing another page out of AMD’s playbook.
Intel launched a whole lot of new Xeon Scalable processors earlier this month, but they neglected to mention a unique line: the U series of single-socket processors. The folks over at Serve The Home sniffed it out first, and Intel has confirmed the existence of the line, just that they “didn’t broadly promote them.”
**[ Read also:[Intel makes a play for high-speed fiber networking for data centers][2] ]**
To backtrack a bit, AMD made a major push for [single-socket servers][3] when it launched the Epyc line of server chips. Epyc comes with up to 32 cores and multithreading, and Intel (and Dell) argued that one 32-core/64-thread processor was enough to handle many loads and a lot cheaper than a two-socket system.
The new U series isn’t available in the regular Intel [ARK database][4] listing of Xeon Scalable processors, but they do show up if you search. Intel says they are looking into that .There are two processors for now, one with 24 cores and two with 20 cores.
The 24-core Intel [Xeon Gold 6212U][5] will be a counterpart to the Intel Xeon Platinum 8260, with a 2.4GHz base clock speed and a 3.9GHz turbo clock and the ability to access up to 1TB of memory. The Xeon Gold 6212U will have the same 165W TDP as the 8260 line, but with a single socket that’s 165 fewer watts of power.
Also, Intel is suggesting a price of about $2,000 for the Intel Xeon Gold 6212U, a big discount over the Xeon Platinum 8260’s $4,702 list price. So, that will translate into much cheaper servers.
The [Intel Xeon Gold 6210U][6] with 20 cores carries a suggested price of $1,500, has a base clock rate of 2.50GHz with turbo boost to 3.9GHz and a 150 watt TDP. Finally, there is the 20-core Intel [Xeon Gold 6209U][7] with a price of around $1,000 that is identical to the 6210 except its base clock speed is 2.1GHz with a turbo boost of 3.9GHz and a TDP of 125 watts due to its lower clock speed.
**[[Get certified as an Apple Technical Coordinator with this seven-part online course from PluralSight.][8] ]**
All of the processors support up to 1TB of DDR4-2933 memory and Intel’s Optane persistent memory.
In terms of speeds and feeds, AMD has a slight advantage over Intel in the single-socket race, and Epyc 2 is rumored to be approaching completion, which will only further advance AMD’s lead.
Join the Network World communities on [Facebook][9] and [LinkedIn][10] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3390201/intel-follows-amds-lead-again-into-single-socket-xeon-servers.html#tk.rss_all
作者：[Andy Patrizio][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Andy-Patrizio/
[b]: https://github.com/lujun9972
[1]: https://images.idgesg.net/images/article/2018/06/intel_generic_cpu_background-100760187-large.jpg
[2]: https://www.networkworld.com/article/3307852/intel-makes-a-play-for-high-speed-fiber-networking-for-data-centers.html
[3]: https://www.networkworld.com/article/3253626/amd-lands-dell-as-its-latest-epyc-server-processor-customer.html
[4]: https://ark.intel.com/content/www/us/en/ark/products/series/192283/2nd-generation-intel-xeon-scalable-processors.html
[5]: https://ark.intel.com/content/www/us/en/ark/products/192453/intel-xeon-gold-6212u-processor-35-75m-cache-2-40-ghz.html
[6]: https://ark.intel.com/content/www/us/en/ark/products/192452/intel-xeon-gold-6210u-processor-27-5m-cache-2-50-ghz.html
[7]: https://ark.intel.com/content/www/us/en/ark/products/193971/intel-xeon-gold-6209u-processor-27-5m-cache-2-10-ghz.html
[8]: https://pluralsight.pxf.io/c/321564/424552/7490?u=https%3A%2F%2Fwww.pluralsight.com%2Fpaths%2Fapple-certified-technical-trainer-10-11
[9]: https://www.facebook.com/NetworkWorld/
[10]: https://www.linkedin.com/company/network-world

69
sources/talk/20190424 IoT roundup- VMware, Nokia beef up their IoT.md
View File

@ -1,69 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (IoT roundup: VMware, Nokia beef up their IoT)
[#]: via: (https://www.networkworld.com/article/3390682/iot-roundup-vmware-nokia-beef-up-their-iot.html#tk.rss_all)
[#]: author: (Jon Gold https://www.networkworld.com/author/Jon-Gold/)
IoT roundup: VMware, Nokia beef up their IoT
======
Everyone wants in on the ground floor of the internet of things, and companies including Nokia, VMware and Silicon Labs are sharpening their offerings in anticipation of further growth.
![Getty Images][1]
When attempting to understand the world of IoT, it’s easy to get sidetracked by all the fascinating use cases: Automated oil and gas platforms! Connected pet feeders! Internet-enabled toilets! (Is “the Internet of Toilets” a thing yet?) But the most important IoT trend to follow may be the way that major tech vendors are vying to make large portions of the market their own.
VMware’s play for a significant chunk of the IoT market is called Pulse IoT Center, and the company released version 2.0 of it this week. It follows the pattern set by other big companies getting into IoT: Leveraging their existing technological strengths and applying them to the messier, more heterodox networking environment that IoT represents.
Unsurprisingly, given that it’s VMware we’re talking about, there’s now a SaaS option, and the company was also eager to talk up that Pulse IoT Center 2.0 has simplified device-onboarding and centralized management features.
**More about edge networking**
* [How edge networking and IoT will reshape data centers][2]
* [Edge computing best practices][3]
* [How edge computing can help secure the IoT][4]
That might sound familiar, and for good reason – companies with any kind of a background in network management, from HPE/Aruba to Amazon, have been pushing to promote their system as the best framework for managing a complicated and often decentralized web of IoT devices from a single platform. By rolling features like software updates, onboarding and security into a single-pane-of-glass management console, those companies are hoping to be the organizational base for customers trying to implement IoT.
Whether they’re successful or not remains to be seen. While major IT companies have been trying to capture market share by competing across multiple verticals, the operational orientation of the IoT also means that non-traditional tech vendors with expertise in particular fields (particularly industrial and automotive) are suddenly major competitors.
**Nokia spreads the IoT network wide**
As a giant carrier-equipment vendor, Nokia is an important company in the overall IoT landscape. While some types of enterprise-focused IoT are heavily localized, like connected factory floors or centrally managed office buildings, others are so geographically disparate that carrier networks are the only connectivity medium that makes sense.
The Finnish company earlier this month broadened its footprint in the IoT space, announcing that it had partnered with Nordic Telecom to create a wide-area network focused on enabling IoT and emergency services. The network, which Nokia is billing as the first mission-critical communications network, operates using LTE technology in the 410-430MHz band – a relatively low frequency, which allows for better propagation and a wide effective range.
The idea is to provide a high-throughput, low-latency network option to any user on the network, whether it’s an emergency services provider needing high-speed video communication or an energy or industrial company with a low-delay-tolerance application.
**Silicon Labs packs more onto IoT chips**
The heart of any IoT implementation remains the SoCs that make devices intelligent in the first place, and Silicon Labs announced that it's building more muscle into its IoT-focused product lineup.
The Austin-based chipmaker said that version 2 of its Wireless Gecko platform will pack more than double the wireless connectivity range of previous entries, which could seriously ease design requirements for companies planning out IoT deployments. The chipsets support Zigbee, Thread and Bluetooth mesh networking, and are designed for line-powered IoT devices, using Arm Cortex-M33 processors for relatively strong computing capacity and high energy efficiency.
Chipset advances aren’t the type of thing that will pay off immediately in terms of making IoT devices more capable, but improvements like these make designing IoT endpoints for particular uses that much easier, and new SoCs will begin to filter into line-of-business equipment over time.
Join the Network World communities on [Facebook][5] and [LinkedIn][6] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3390682/iot-roundup-vmware-nokia-beef-up-their-iot.html#tk.rss_all
作者：[Jon Gold][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Jon-Gold/
[b]: https://github.com/lujun9972
[1]: https://images.idgesg.net/images/article/2018/08/nw_iot-news_internet-of-things_smart-city_smart-home5-100768494-large.jpg
[2]: https://www.networkworld.com/article/3291790/data-center/how-edge-networking-and-iot-will-reshape-data-centers.html
[3]: https://www.networkworld.com/article/3331978/lan-wan/edge-computing-best-practices.html
[4]: https://www.networkworld.com/article/3331905/internet-of-things/how-edge-computing-can-help-secure-the-iot.html
[5]: https://www.facebook.com/NetworkWorld/
[6]: https://www.linkedin.com/company/network-world

52
sources/talk/20190425 Dell EMC and Cisco renew converged infrastructure alliance.md
View File

@ -1,52 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Dell EMC and Cisco renew converged infrastructure alliance)
[#]: via: (https://www.networkworld.com/article/3391071/dell-emc-and-cisco-renew-converged-infrastructure-alliance.html#tk.rss_all)
[#]: author: (Andy Patrizio https://www.networkworld.com/author/Andy-Patrizio/)
Dell EMC and Cisco renew converged infrastructure alliance
======
Dell EMC and Cisco renewed their agreement to collaborate on converged infrastructure (CI) products for a few more years even though the momentum is elsewhere.
![Dell EMC][1]
Dell EMC and Cisco have renewed a collaboration on converged infrastructure (CI) products that has run for more than a decade, even as the momentum shifts elsewhere. The news was announced via a [blog post][2] by Pete Manca, senior vice president for solutions engineering at Dell EMC.
The deal is centered around Dell EMC’s VxBlock product line, which originally started out in 2009 as a joint venture between EMC and Cisco called VCE (Virtual Computing Environment). EMC bought out Cisco’s stake in the venture before Dell bought EMC.
The devices offered UCS servers and networking from Cisco, EMC storage, and VMware virtualization software in pre-configured, integrated bundles. VCE was retired in favor of new brands, VxBlock, VxRail, and VxRack. The lineup has been pared down to one device, the VxBlock 1000.
**[ Read also:[How to plan a software-defined data-center network][3] ]**
“The newly inked agreement entails continued company alignment across multiple organizations: executive, product development, marketing, and sales,” Manca wrote in the blog post. “This means we’ll continue to share product roadmaps and collaborate on strategic development activities, with Cisco investing in a range of Dell EMC sales, marketing and training initiatives to support VxBlock 1000.”
Dell EMC cites IDC research that it holds a 48% market share in converged systems, nearly 1.5 times that of any other vendor. But IDC's April edition of the Converged Systems Tracker said the CI category is on the downswing. CI sales fell 6.4% year over year, while the market for hyperconverged infrastructure (HCI) grew 57.2% year over year.
For the unfamiliar, the primary difference between converged and hyperconverged infrastructure is that CI relies on hardware building blocks, while HCI is software-defined and considered more flexible and scalable than CI and operates more like a cloud system with resources spun up and down as needed.
Despite this, Dell is committed to CI systems. Just last month it announced an update and expansion of the VxBlock 1000, including higher scalability, a broader choice of components, and the option to add new technologies. It featured updated VMware vRealize and vSphere support, the option to consolidate high-value, mission-critical workloads with new storage and data protection options and support for Cisco UCS fabric and servers.
For customers who prefer to build their own infrastructure solutions, Dell EMC introduced Ready Stack, a portfolio of validated designs with sizing, design, and deployment resources that offer VMware-based IaaS, vSphere on Dell EMC PowerEdge servers and Dell EMC Unity storage, and Microsoft Hyper-V on Dell EMC PowerEdge servers and Dell EMC Unity storage.
Join the Network World communities on [Facebook][4] and [LinkedIn][5] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3391071/dell-emc-and-cisco-renew-converged-infrastructure-alliance.html#tk.rss_all
作者：[Andy Patrizio][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Andy-Patrizio/
[b]: https://github.com/lujun9972
[1]: https://images.idgesg.net/images/article/2019/04/dell-emc-vxblock-1000-100794721-large.jpg
[2]: https://blog.dellemc.com/en-us/dell-technologies-cisco-reaffirm-joint-commitment-converged-infrastructure/
[3]: https://www.networkworld.com/article/3284352/data-center/how-to-plan-a-software-defined-data-center-network.html
[4]: https://www.facebook.com/NetworkWorld/
[5]: https://www.linkedin.com/company/network-world

86
sources/talk/20190429 Venerable Cisco Catalyst 6000 switches ousted by new Catalyst 9600.md
View File

@ -1,86 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Venerable Cisco Catalyst 6000 switches ousted by new Catalyst 9600)
[#]: via: (https://www.networkworld.com/article/3391580/venerable-cisco-catalyst-6000-switches-ousted-by-new-catalyst-9600.html#tk.rss_all)
[#]: author: (Michael Cooney https://www.networkworld.com/author/Michael-Cooney/)
Venerable Cisco Catalyst 6000 switches ousted by new Catalyst 9600
======
Cisco introduced Catalyst 9600 switches, that let customers automate, set policy, provide security and gain assurance across wired and wireless networks.
![Martyn Williams][1]
Few events in the tech industry are truly transformative, but Cisco’s replacement of its core Catalyst 6000 family could be one of those actions for customers and the company.
Introduced in 1999, [iterations of the Catalyst 6000][2] have nestled into the core of scores of enterprise networks, with the model 6500 becoming the company’s largest selling box ever.
**Learn about edge networking**
* [How edge networking and IoT will reshape data centers][3]
* [Edge computing best practices][4]
* [How edge computing can help secure the IoT][5]
It goes without question that migrating these customers alone to the new switch – the Catalyst 9600 which the company introduced today – will be of monumental importance to Cisco as it looks to revamp and continue to dominate large campus-core deployments. The first [Catalyst 9000][6], introduced in June 2017, is already the fastest ramping product line in Cisco’s history.
“There are at least tens of thousands of Cat 6000s running in campus cores all over the world,” said [Sachin Gupta][7], senior vice president for product management at Cisco. ”It is the Swiss Army knife of switches in term of features, and we have taken great care and over two years developing feature parity and an easy migration path for those users to the Cat 9000.”
Indeed the 9600 brings with it for Cat 6000 features such as support for MPLS, virtual switching and IPv6, while adding or bolstering support for newer items such as Intent-based networking (IBN), wireless networks and security segmentation. Strategically the 9600 helps fill out the company’s revamped lineup which includes the 9200 family of access switches, the [9500][8] aggregation switch and [9800 wireless controller.][9]
Some of the nitty-gritty details about the 9600:
* It is a purpose-built 40 Gigabit and 100 Gigabit Ethernet line of modular switches targeted for the enterprise campus with a wired switching capacity of up to 25.6 Tbps, with up to 6.4 Tbps of bandwidth per slot.
* The switch supports granular port densities that fit diverse campus needs, including nonblocking 40 Gigabit and 100 Gigabit Ethernet Quad Small Form-Factor Pluggable (QSFP+, QSFP28) and 1, 10, and 25 GE Small Form-Factor Pluggable Plus (SFP, SFP+, SFP28).
* It can be configured to support up to 48 nonblocking 100 Gigabit Ethernet QSPF28 ports with the Cisco Catalyst 9600 Series Supervisor Engine 1; Up to 96 nonblocking 40 Gigabit Ethernet QSFP+ ports with the Cisco Catalyst 9600 Series Supervisor Engine 1 and Up to 192 nonblocking 25 Gigabit/10 Gigabit Ethernet SFP28/SFP+ ports with the Cisco Catalyst 9600 Series Supervisor Engine 1.
* It supports advanced routing and infrastructure services (MPLS, Layer 2 and Layer 3 VPNs, Multicast VPN, and Network Address Translation.
* Cisco Software-Defined Access capabilities (such as a host-tracking database, cross-domain connectivity, and VPN Routing and Forwarding [VRF]-aware Locator/ID Separation Protocol; and network system virtualization with Cisco StackWise virtual technology.
The 9600 series runs Cisco’s IOS XE software which now runs across all Catalyst 9000 family members. The software brings with it support for other key products such as Cisco’s [DNA Center][10] which controls automation capabilities, assurance setting, fabric provisioning and policy-based segmentation for enterprise networks. What that means is that with one user interface, DNA Center, customers can automate, set policy, provide security and gain assurance across the entire wired and wireless network fabric, Gupta said.
“The 9600 is a big deal for Cisco and customers as it brings together the campus core and lets users establish standards access and usage policies across their wired and wireless environments,” said Brandon Butler, a senior research analyst with IDC. “It was important that Cisco add a powerful switch to handle the increasing amounts of traffic wireless and cloud applications are bringing to the network.”
IOS XE brings with it automated device provisioning and a wide variety of automation features including support for the network configuration protocol NETCONF and RESTCONF using YANG data models. The software offers near-real-time monitoring of the network, leading to quick detection and rectification of failures, Cisco says.
The software also supports hot patching which provides fixes for critical bugs and security vulnerabilities between regular maintenance releases. This support lets customers add patches without having to wait for the next maintenance release, Cisco says.
As with the rest of the Catalyst family, the 9600 is available via subscription-based licensing. Cisco says the [base licensing package][11] includes Network Advantage licensing options that are tied to the hardware. The base licensing packages cover switching fundamentals, management automation, troubleshooting, and advanced switching features. These base licenses are perpetual.
An add-on licensing package includes the Cisco DNA Premier and Cisco DNA Advantage options. The Cisco DNA add-on licenses are available as a subscription.
IDC’S Butler noted that there are competitors such as Ruckus, Aruba and Extreme that offer switches capable of melding wired and wireless environments.
The new switch is built for the next two decades of networking, Gupta said. “If any of our competitors though they could just go in and replace the Cat 6k they were misguided.”
Join the Network World communities on [Facebook][12] and [LinkedIn][13] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3391580/venerable-cisco-catalyst-6000-switches-ousted-by-new-catalyst-9600.html#tk.rss_all
作者：[Michael Cooney][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Michael-Cooney/
[b]: https://github.com/lujun9972
[1]: https://images.techhive.com/images/article/2017/02/170227-mwc-02759-100710709-large.jpg
[2]: https://www.networkworld.com/article/2289826/133715-The-illustrious-history-of-Cisco-s-Catalyst-LAN-switches.html
[3]: https://www.networkworld.com/article/3291790/data-center/how-edge-networking-and-iot-will-reshape-data-centers.html
[4]: https://www.networkworld.com/article/3331978/lan-wan/edge-computing-best-practices.html
[5]: https://www.networkworld.com/article/3331905/internet-of-things/how-edge-computing-can-help-secure-the-iot.html
[6]: https://www.networkworld.com/article/3256264/cisco-ceo-we-are-still-only-on-the-front-end-of-a-new-version-of-the-network.html
[7]: https://blogs.cisco.com/enterprise/looking-forward-catalyst-9600-switch-and-9100-access-point-meraki
[8]: https://www.networkworld.com/article/3202105/cisco-brings-intent-based-networking-to-the-end-to-end-network.html
[9]: https://www.networkworld.com/article/3321000/cisco-links-wireless-wired-worlds-with-new-catalyst-9000-switches.html
[10]: https://www.networkworld.com/article/3280988/cisco/cisco-opens-dna-center-network-control-and-management-software-to-the-devops-masses.html
[11]: https://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst9600/software/release/16-11/release_notes/ol-16-11-9600.html#id_67835
[12]: https://www.facebook.com/NetworkWorld/
[13]: https://www.linkedin.com/company/network-world

59
sources/tech/20190327 HPE introduces hybrid cloud consulting business.md
View File

@ -1,59 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (HPE introduces hybrid cloud consulting business)
[#]: via: (https://www.networkworld.com/article/3384919/hpe-introduces-hybrid-cloud-consulting-business.html#tk.rss_all)
[#]: author: (Andy Patrizio https://www.networkworld.com/author/Andy-Patrizio/)
HPE introduces hybrid cloud consulting business
======
### HPE's Right Mix Advisor is designed to find a balance between on-premises and cloud systems.
![Hewlett Packard Enterprise][1]
Hybrid cloud is pretty much the de facto way to go, with only a few firms adopting a pure cloud play to replace their data center and only suicidal firms refusing to go to the cloud. But picking the right balance between on-premises and the cloud is tricky, and a mistake can be costly.
Enter Right Mix Advisor from Hewlett Packard Enterprise, a combination of consulting from HPE's Pointnext division and software tools. It draws on quite a bit of recent acquisitions. Another part of Right Mix Advisor is a British cloud consultancy RedPixie, Amazon Web Services (AWS) specialists Cloud Technology Partners, and automated discovery capabilities from an Irish startup iQuate.
Right Mix Advisor gathers data points from the company’s entire enterprise, ranging from configuration management database systems (CMDBs), such as ServiceNow, to external sources, such as cloud providers. HPE says that in a recent customer engagement it scanned 9 million IP addresses across six data centers.
**[ Read also:[What is hybrid cloud computing][2]. | Learn [what you need to know about multi-cloud][3]. | Get regularly scheduled insights by [signing up for Network World newsletters][4]. ]**
HPE Pointnext consultants then work with the client’s IT teams to analyze the data to determine the optimal configuration for workload placement. Pointnext has become HPE’s main consulting outfit following its divestiture of EDS, which it acquired in 2008 but spun off in a merger with CSC to form DXC Consulting. Pointnext now has 25,000 consultants in 88 countries.
In a typical engagement, HPE claims it can deliver a concrete action plan within weeks, whereas previously businesses may have needed months to come to a conclusion using a manual processes. HPE has found migrating the right workloads to the right mix of hybrid cloud can typically result in 40 percent total cost of ownership savings*. *
Although HPE has thrown its weight behind AWS, that doesn’t mean it doesn’t support competitors. Erik Vogel, vice president of hybrid IT for HPE Pointnext, notes in the blog post announcing Right Mix Advisor that target environments could be Microsoft Azure or Azure Stack, AWS, Google or Ali Cloud.
“New service providers are popping up every day, and we see the big public cloud providers constantly producing new services and pricing models. As a result, the calculus for determining your right mix is constantly changing. If Azure, for example, offers a new service capability or a 10 percent pricing discount and it makes sense to leverage it, you want to be able to move an application seamlessly into that new environment,” he wrote.
Key to Right Mix Advisor is app migration, and Pointnext follows the 50/30/20 rule: About 50 percent of apps are suitable for migration to the cloud, and for about 30 percent, migration is not a good choice for migration to be worth the effort. The remaining 20 percent should be retired.
“With HPE Right Mix Advisor, you can identify that 50 percent,” he wrote. “Rather than hand you a laundry list of 10,000 apps to start migrating, HPE Right Mix Advisor hones in on what’s most impactful right now to meet your business goals – the 10 things you can do on Monday morning that you can be confident will really help your business.”
HPE has already done some pilot projects with the Right Mix service and expects to expand it to include channel partners.
Join the Network World communities on [Facebook][5] and [LinkedIn][6] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3384919/hpe-introduces-hybrid-cloud-consulting-business.html#tk.rss_all
作者：[Andy Patrizio][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Andy-Patrizio/
[b]: https://github.com/lujun9972
[1]: https://images.techhive.com/images/article/2015/11/hpe_building-100625424-large.jpg
[2]: https://www.networkworld.com/article/3233132/cloud-computing/what-is-hybrid-cloud-computing.html
[3]: https://www.networkworld.com/article/3252775/hybrid-cloud/multicloud-mania-what-to-know.html
[4]: https://www.networkworld.com/newsletters/signup.html
[5]: https://www.facebook.com/NetworkWorld/
[6]: https://www.linkedin.com/company/network-world

72
sources/tech/20190328 Cisco warns of two security patches that don-t work, issues 17 new ones for IOS flaws.md
View File

@ -1,72 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Cisco warns of two security patches that don’t work, issues 17 new ones for IOS flaws)
[#]: via: (https://www.networkworld.com/article/3384742/cisco-warns-of-two-security-patches-that-dont-work-issues-17-new-ones-for-ios-flaws.html#tk.rss_all)
[#]: author: (Michael Cooney https://www.networkworld.com/author/Michael-Cooney/)
Cisco warns of two security patches that don’t work, issues 17 new ones for IOS flaws
======
### Cisco is issuing 17 new fixes for security problems with IOS and IOS/XE software that runs most of its routers and switches, while it has no patch yet to replace flawed patches to RV320 and RV 325 routers.
![Marisa9 / Getty][1]
Cisco has dropped [17 Security advisories describing 19 vulnerabilities][2] in the software that runs most of its routers and switches, IOS and IOS/XE.
The company also announced that two previously issued patches for its RV320 and RV325 Dual Gigabit WAN VPN Routers were “incomplete” and would need to be redone and reissued.
**[ Also see[What to consider when deploying a next generation firewall][3]. | Get regularly scheduled insights by [signing up for Network World newsletters][4]. ]**
Cisco rates both those router vulnerabilities as “High” and describes the problems like this:
* [One vulnerability][5] is due to improper validation of user-supplied input. An attacker could exploit this vulnerability by sending malicious HTTP POST requests to the web-based management interface of an affected device. A successful exploit could allow the attacker to execute arbitrary commands on the underlying Linux shell as _root_.
* The [second exposure][6] is due to improper access controls for URLs. An attacker could exploit this vulnerability by connecting to an affected device via HTTP or HTTPS and requesting specific URLs. A successful exploit could allow the attacker to download the router configuration or detailed diagnostic information.
Cisco said firmware updates that address these vulnerabilities are not available and no workarounds exist, but is working on a complete fix for both.
On the IOS front, the company said six of the vulnerabilities affect both Cisco IOS Software and Cisco IOS XE Software, one of the vulnerabilities affects just Cisco IOS software and ten of the vulnerabilities affect just Cisco IOS XE software. Some of the security bugs, which are all rated as “High”, include:
* [A vulnerability][7] in the web UI of Cisco IOS XE Software could let an unauthenticated, remote attacker access sensitive configuration information.
* [A vulnerability][8] in Cisco IOS XE Software could let an authenticated, local attacker inject arbitrary commands that are executed with elevated privileges. The vulnerability is due to insufficient input validation of commands supplied by the user. An attacker could exploit this vulnerability by authenticating to a device and submitting crafted input to the affected commands.
* [A weakness][9] in the ingress traffic validation of Cisco IOS XE Software for Cisco Aggregation Services Router (ASR) 900 Route Switch Processor 3 could let an unauthenticated, adjacent attacker trigger a reload of an affected device, resulting in a denial of service (DoS) condition, Cisco said. The vulnerability exists because the software insufficiently validates ingress traffic on the ASIC used on the RSP3 platform. An attacker could exploit this vulnerability by sending a malformed OSPF version 2 message to an affected device.
* A problem in the [authorization subsystem][10] of Cisco IOS XE Software could allow an authenticated but unprivileged (level 1), remote attacker to run privileged Cisco IOS commands by using the web UI. The vulnerability is due to improper validation of user privileges of web UI users. An attacker could exploit this vulnerability by submitting a malicious payload to a specific endpoint in the web UI, Cisco said.
* A vulnerability in the [Cluster Management Protocol][11] (CMP) processing code in Cisco IOS Software and Cisco IOS XE Software could allow an unauthenticated, adjacent attacker to trigger a DoS condition on an affected device. The vulnerability is due to insufficient input validation when processing CMP management packets, Cisco said.
Cisco has released free software updates that address the vulnerabilities described in these advisories and [directs users to their software agreements][12] to find out how they can download the fixes.
Join the Network World communities on [Facebook][13] and [LinkedIn][14] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3384742/cisco-warns-of-two-security-patches-that-dont-work-issues-17-new-ones-for-ios-flaws.html#tk.rss_all
作者：[Michael Cooney][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Michael-Cooney/
[b]: https://github.com/lujun9972
[1]: https://images.idgesg.net/images/article/2019/02/woman-with-hands-over-face_mistake_oops_embarrassed_shy-by-marisa9-getty-100787990-large.jpg
[2]: https://tools.cisco.com/security/center/viewErp.x?alertId=ERP-71135
[3]: https://www.networkworld.com/article/3236448/lan-wan/what-to-consider-when-deploying-a-next-generation-firewall.html
[4]: https://www.networkworld.com/newsletters/signup.html
[5]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190123-rv-inject
[6]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190123-rv-info
[7]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190327-xeid
[8]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190327-xecmd
[9]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190327-rsp3-ospf
[10]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190327-iosxe-privesc
[11]: https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20190327-cmp-dos
[12]: https://www.cisco.com/c/en/us/about/legal/cloud-and-software/end_user_license_agreement.html
[13]: https://www.facebook.com/NetworkWorld/
[14]: https://www.linkedin.com/company/network-world

90
sources/tech/20190402 Announcing the release of Fedora 30 Beta.md
View File

@ -1,90 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (Announcing the release of Fedora 30 Beta)
[#]: via: (https://fedoramagazine.org/announcing-the-release-of-fedora-30-beta/)
[#]: author: (Ben Cotton https://fedoramagazine.org/author/bcotton/)
Announcing the release of Fedora 30 Beta
======
![][1]
The Fedora Project is pleased to announce the immediate availability of Fedora 30 Beta, the next big step on our journey to the exciting Fedora 30 release.
Download the prerelease from our Get Fedora site:
* [Get Fedora 30 Beta Workstation][2]
* [Get Fedora 30 Beta Server][3]
* [Get Fedora 30 Beta Silverblue][4]
Or, check out one of our popular variants, including KDE Plasma, Xfce, and other desktop environments, as well as images for ARM devices like the Raspberry Pi 2 and 3:
* [Get Fedora 30 Beta Spins][5]
* [Get Fedora 30 Beta Labs][6]
* [Get Fedora 30 Beta ARM][7]
### Beta Release Highlights
#### New desktop environment options
Fedora 30 Beta includes two new options for desktop environment. [DeepinDE][8] and [Pantheon Desktop][9] join GNOME, KDE Plasma, Xfce, and others as options for users to customize their Fedora experience.
#### DNF performance improvements
All dnf repository metadata for Fedora 30 Beta is compressed with the zchunk format in addition to xz or gzip. zchunk is a new compression format designed to allow for highly efficient deltas. When Fedora’s metadata is compressed using zchunk, dnf will download only the differences between any earlier copies of the metadata and the current version.
#### GNOME 3.32
Fedora 30 Workstation Beta includes GNOME 3.32, the latest version of the popular desktop environment. GNOME 3.32 features updated visual style, including the user interface, the icons, and the desktop itself. For a full list of GNOME 3.32 highlights, see the [release notes][10].
#### Other updates
Fedora 30 Beta also includes updated versions of many popular packages like Golang, the Bash shell, the GNU C Library, Python, and Perl. For a full list, see the [Change set][11] on the Fedora Wiki. In addition, many Python 2 packages are removed in preparation for Python 2 end-of-life on 2020-01-01.
#### Testing needed
Since this is a Beta release, we expect that you may encounter bugs or missing features. To report issues encountered during testing, contact the Fedora QA team via the mailing list or in #fedora-qa on Freenode. As testing progresses, common issues are tracked on the [Common F30 Bugs page][12].
For tips on reporting a bug effectively, read [how to file a bug][13].
#### What is the Beta Release?
A Beta release is code-complete and bears a very strong resemblance to the final release. If you take the time to download and try out the Beta, you can check and make sure the things that are important to you are working. Every bug you find and report doesn’t just help you, it improves the experience of millions of Fedora users worldwide! Together, we can make Fedora rock-solid. We have a culture of coordinating new features and pushing fixes upstream as much as we can. Your feedback improves not only Fedora, but Linux and free software as a whole.
#### More information
For more detailed information about what’s new on Fedora 30 Beta release, you can consult the [Fedora 30 Change set][11]. It contains more technical information about the new packages and improvements shipped with this release.
--------------------------------------------------------------------------------
via: https://fedoramagazine.org/announcing-the-release-of-fedora-30-beta/
作者：[Ben Cotton][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://fedoramagazine.org/author/bcotton/
[b]: https://github.com/lujun9972
[1]: https://fedoramagazine.org/wp-content/uploads/2019/03/f30-beta-816x345.jpg
[2]: https://getfedora.org/workstation/prerelease/
[3]: https://getfedora.org/server/prerelease/
[4]: https://silverblue.fedoraproject.org/download
[5]: https://spins.fedoraproject.org/prerelease
[6]: https://labs.fedoraproject.org/prerelease
[7]: https://arm.fedoraproject.org/prerelease
[8]: https://www.deepin.org/en/dde/
[9]: https://www.fosslinux.com/4652/pantheon-everything-you-need-to-know-about-the-elementary-os-desktop.htm
[10]: https://help.gnome.org/misc/release-notes/3.32/
[11]: https://fedoraproject.org/wiki/Releases/30/ChangeSet
[12]: https://fedoraproject.org/wiki/Common_F30_bugs
[13]: https://docs.fedoraproject.org/en-US/quick-docs/howto-file-a-bug/

70
sources/tech/20190403 How to rebase to Fedora 30 Beta on Silverblue.md
View File

@ -1,70 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (How to rebase to Fedora 30 Beta on Silverblue)
[#]: via: (https://fedoramagazine.org/how-to-rebase-to-fedora-30-beta-on-silverblue/)
[#]: author: (Michal Konečný https://fedoramagazine.org/author/zlopez/)
How to rebase to Fedora 30 Beta on Silverblue
======
![][1]
Silverblue is [an operating system for your desktop built on Fedora][2]. It’s excellent for daily use, development, and container-based workflows. It offers [numerous advantages][3] such as being able to roll back in case of any problems. If you want to test Fedora 30 on your Silverblue system, this article tells you how. It not only shows you what to do, but also how to revert back if anything unforeseen happens.
### Switching to Fedora 30 branch
Switching to Fedora 30 on Silverblue is easy. First, check if the _30_ branch is available, which should be true now:
```
ostree remote refs fedora-workstation
```
You should see the following in the output:
```
fedora-workstation:fedora/30/x86_64/silverblue
```
Next, import the GPG key for the Fedora 30 branch. Without this step, you won’t be able to rebase.
```
sudo ostree remote gpg-import fedora-workstation -k /etc/pki/rpm-gpg/RPM-GPG-KEY-fedora-30-primary
```
Next, rebase your system to the Fedora 30 branch.
```
rpm-ostree rebase fedora-workstation:fedora/30/x86_64/silverblue
```
Finally, the last thing to do is restart your computer and boot to Fedora 30.
### How to revert things back
Remember that Fedora 30’s still in beta testing phase, so there could still be some issues. If anything bad happens — for instance, if you can’t boot to Fedora 30 at all — it’s easy to go back. Just pick the previous entry in GRUB, and your system will start in its previous state before switching to Fedora 30. To make this change permanent, use the following command:
```
rpm-ostree rollback
```
That’s it. Now you know how to rebase to Fedora 30 and back. So why not test it today? 🙂
--------------------------------------------------------------------------------
via: https://fedoramagazine.org/how-to-rebase-to-fedora-30-beta-on-silverblue/
作者：[Michal Konečný][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://fedoramagazine.org/author/zlopez/
[b]: https://github.com/lujun9972
[1]: https://fedoramagazine.org/wp-content/uploads/2019/03/silverblue-f30beta-816x345.jpg
[2]: https://docs.fedoraproject.org/en-US/fedora-silverblue/
[3]: https://fedoramagazine.org/give-fedora-silverblue-a-test-drive/

69
sources/tech/20190409 The Microsoft-BMW IoT Open Manufacturing Platform might not be so open.md
View File

@ -1,69 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (The Microsoft/BMW IoT Open Manufacturing Platform might not be so open)
[#]: via: (https://www.networkworld.com/article/3387642/the-microsoftbmw-iot-open-manufacturing-platform-might-not-be-so-open.html#tk.rss_all)
[#]: author: (Fredric Paul https://www.networkworld.com/author/Fredric-Paul/)
The Microsoft/BMW IoT Open Manufacturing Platform might not be so open
======
The new industrial IoT Open Manufacturing Platform from Microsoft and BMW runs only on Microsoft Azure. That could be an issue.
![Martyn Williams][1]
Last week at [Hannover Messe][2], Microsoft and German carmaker BMW announced a partnership to build a hardware and software technology framework and reference architecture for the industrial internet of things (IoT), and foster a community to spread these smart-factory solutions across the automotive and manufacturing industries.
The stated goal of the [Open Manufacturing Platform (OMP)][3]? According to the press release, it's “to drive open industrial IoT development and help grow a community to build future [Industry 4.0][4] solutions.” To make that a reality, the companies said that by the end of 2019, they plan to attract four to six partners — including manufacturers and suppliers from both inside and outside the automotive industry — and to have rolled out at least 15 use cases operating in actual production environments.
**[ Read also:[An inside look at an IIoT-powered smart factory][5] | Get regularly scheduled insights: [Sign up for Network World newsletters][6] ]**
### Complex and proprietary is bad for IoT
It sounds like a great idea, right? As the companies rightly point out, many of today’s industrial IoT solutions rely on “complex, proprietary systems that create data silos and slow productivity.” Who wouldn’t want to “standardize data models that enable analytics and machine learning scenarios” and “accelerate future industrial IoT developments, shorten time to value, and drive production efficiencies while addressing common industrial challenges”?
But before you get too excited, let’s talk about a key word in the effort: open. As Scott Guthrie, executive vice president of Microsoft Cloud + AI Group, said in a statement, "Our commitment to building an open community will create new opportunities for collaboration across the entire manufacturing value chain."
### The Open Manufacturing Platform is open only to Microsoft Azure
However, that will happen as long as all that collaboration occurs in Microsoft Azure. I’m not saying Azure isn’t up to the task, but it’s hardly the only (or even the leading) cloud platform interested in the industrial IoT. Putting everything in Azure might be an issue to those potential OMP partners. It’s an “open” question as to how many companies already invested in Amazon Web Services (AWS) or the Google Cloud Platform (GCP) will be willing to make the switch or go multi-cloud just to take advantage of the OMP.
My guess is that Microsoft and BMW won’t have too much trouble meeting their initial goals for the OMP. It shouldn’t be that hard to get a handful of existing Azure customers to come up with 15 use cases leveraging advances in analytics, artificial intelligence (AI), and digital feedback loops. (As an example, the companies cited the autonomous transport systems in BMW’s factory in Regensburg, Germany, part of the more than 3,000 machines, robots and transport systems connected with the BMW Group’s IoT platform, which — naturally — is built on Microsoft Azure's cloud.)
### Will non-Azure users jump on board the OMP?
The question is whether tying all this to a single cloud provider will affect the effort to attract enough new companies — including companies not currently using Azure — to establish a truly viable open platform?
Perhaps [Stacey Higginbotham at Stacy on IoT put it best][7]:
> “What they really launched is a reference design for manufacturers to work from.”
That’s not nothing, of course, but it’s a lot less ambitious than building a new industrial IoT platform. And it may not easily fulfill the vision of a community working together to create shared solutions that benefit everyone.
**[ Now read this:[Why are IoT platforms so darn confusing?][8] ]**
Join the Network World communities on [Facebook][9] and [LinkedIn][10] to comment on topics that are top of mind.
--------------------------------------------------------------------------------
via: https://www.networkworld.com/article/3387642/the-microsoftbmw-iot-open-manufacturing-platform-might-not-be-so-open.html#tk.rss_all
作者：[Fredric Paul][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://www.networkworld.com/author/Fredric-Paul/
[b]: https://github.com/lujun9972
[1]: https://images.techhive.com/images/article/2017/01/20170107_105344-100702818-large.jpg
[2]: https://www.hannovermesse.de/home
[3]: https://www.prnewswire.co.uk/news-releases/microsoft-and-the-bmw-group-launch-the-open-manufacturing-platform-859672858.html
[4]: https://en.wikipedia.org/wiki/Industry_4.0
[5]: https://www.networkworld.com/article/3384378/an-inside-look-at-tempo-automations-iiot-powered-smart-factory.html
[6]: https://www.networkworld.com/newsletters/signup.html
[7]: https://mailchi.mp/iotpodcast/stacey-on-iot-industrial-iot-reminds-me-of-apples-ecosystem?e=6bf9beb394
[8]: https://www.networkworld.com/article/3336166/why-are-iot-platforms-so-darn-confusing.html
[9]: https://www.facebook.com/NetworkWorld/
[10]: https://www.linkedin.com/company/network-world

115
sources/tech/20190430 The Awesome Fedora 30 is Here- Check Out the New Features.md
View File

@ -1,115 +0,0 @@
[#]: collector: (lujun9972)
[#]: translator: ( )
[#]: reviewer: ( )
[#]: publisher: ( )
[#]: url: ( )
[#]: subject: (The Awesome Fedora 30 is Here! Check Out the New Features)
[#]: via: (https://itsfoss.com/fedora-30/)
[#]: author: (Abhishek Prakash https://itsfoss.com/author/abhishek/)
The Awesome Fedora 30 is Here! Check Out the New Features
======
The latest and greatest release of Fedora is here. Fedora 30 brings some visual as well as performance improvements.
Fedora releases a new version every six months and each release is supported for thirteen months.
Before you decide to download or upgrade Fedora, let’s first see what’s new in Fedora 30.
### New Features in Fedora 30
![Fedora 30 Release][1]
Here’s what’s new in the latest release of Fedora.
#### GNOME 3.32 gives a brand new look, features and performance improvement
A lot of visual improvements is brought by the latest release of GNOME.
GNOME 3.32 has refreshed new icons and UI and it almost looks like a brand new version of GNOME.
![Gnome 3.32 icons | Image Credit][2]
GNOME 3.32 also brings several other features like fractional scaling, permission control for each application, granular control on Night Light intensity among many other changes.
GNOME 3.32 also brings some performance improvements. You’ll see faster file and app searches and a smoother scrolling.
#### Improved performance for DNF
Fedora 30 will see a faster [DNF][3] (the default package manager for Fedora) thanks to the [zchunk][4] compression algorithm.
The zchunk algorithm splits the file into independent chunks. This helps in dealing with ‘delta’ or changes as you download only the changed chunks while downloading the new version of a file.
With zcunk, dnf will only download the difference between the metadata of the current version and the earlier versions.
#### Fedora 30 brings two new desktop environments into the fold
Fedora already offers several desktop environment choices. Fedora 30 extends the offering with [elementary OS][5]‘ Pantheon desktop environment and Deepin Linux’ [DeepinDE][6].
So now you can enjoy the looks and feel of elementary OS and Deepin Linux in Fedora. How cool is that!
#### Linux Kernel 5
Fedora 29 has Linux Kernel 5.0.9 version that has improved support for hardware and some performance improvements. You may check out the [features of Linux kernel 5.0 in this article][7].
[][8]
Suggested read The Featureful Release of Nextcloud 14 Has Two New Security Features
#### Updated software
You’ll also get newer versions of software. Some of the major ones are:
* GCC 9.0.1
* [Bash Shell 5.0][9]
* GNU C Library 2.29
* Ruby 2.6
* Golang 1.12
* Mesa 19.0.2
* Vagrant 2.2
* JDK12
* PHP 7.3
* Fish 3.0
* Erlang 21
* Python 3.7.3
### Getting Fedora 30
If you are already using Fedora 29 then you can upgrade to the latest release from your current install. You may follow this guide to learn [how to upgrade a Fedora version][10].
Fedora 29 users will still get the updates for seven more months so if you don’t feel like upgrading, you may skip it for now. Fedora 28 users have no choice because Fedora 28 reached end of life next month which means there will be no security or maintenance update anymore. Upgrading to a newer version is no longer a choice.
You always has the option to download the ISO of Fedora 30 and install it afresh. You can download Fedora from its official website. It’s only available for 64-bit systems and the ISO is 1.9 GB in size.
[Download Fedora 30 Workstation][11]
What do you think of Fedora 30? Are you planning to upgrade or at least try it out? Do share your thoughts in the comment section.
--------------------------------------------------------------------------------
via: https://itsfoss.com/fedora-30/
作者：[Abhishek Prakash][a]
选题：[lujun9972][b]
译者：[译者ID](https://github.com/译者ID)
校对：[校对者ID](https://github.com/校对者ID)
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
[a]: https://itsfoss.com/author/abhishek/
[b]: https://github.com/lujun9972
[1]: https://itsfoss.com/wp-content/uploads/2019/04/fedora-30-release-800x450.png
[2]: https://itsfoss.com/wp-content/uploads/2019/04/gnome-3-32-icons.png
[3]: https://fedoraproject.org/wiki/DNF?rd=Dnf
[4]: https://github.com/zchunk/zchunk
[5]: https://itsfoss.com/elementary-os-juno-features/
[6]: https://www.deepin.org/en/dde/
[7]: https://itsfoss.com/linux-kernel-5/
[8]: https://itsfoss.com/nextcloud-14-release/
[9]: https://itsfoss.com/bash-5-release/
[10]: https://itsfoss.com/upgrade-fedora-version/
[11]: https://getfedora.org/en/workstation/