mirror of
https://github.com/LCTT/TranslateProject.git
synced 2025-01-22 23:00:57 +08:00
translated
This commit is contained in:
HankChow
parent
0e1722f33b
commit
da3059bfdf
@ -1,81 +0,0 @@
|
||||
HankChow translating
|
||||
|
||||
5 tips for choosing the right open source database
|
||||
======
|
||||
When selecting a mission-critical application, you can't afford to make mistakes.
|
||||
|
||||
|
||||
|
||||
So, your company has a directive to adopt more open source database technologies, and they've recruited you to select the right direction. Whether you are an open source technology veteran or a newcomer, this is a daunting and overwhelming task.
|
||||
|
||||
Over the past several years, open source technology adoption has steadily increased in the enterprise space. With its popularity comes a crowded marketplace with open source software companies promising that their solution will solve every problem and fit every workload. Be wary of these promises. Choosing the right open source technology—especially a database—is an important and difficult decision you can't make lightly.
|
||||
|
||||
In my experience as an IT professional at [Percona][1] and other companies, I've been fortunate to work hands-on in adopting open source technologies and guiding others in making the right decisions. There are many important factors to consider; hopefully, this article will shine a light on a few.
|
||||
|
||||
### 1. Have a goal.
|
||||
|
||||
This may seem simple, but based on my many conversations with people exploring MySQL, MongoDB, or PostgreSQL, it is top of the list in importance.
|
||||
|
||||
To avoid getting overwhelmed by the unlimited combinations of open source database software in the market, have a specific goal in mind. Maybe your goal is to provide your internal developers with a standardized, open source database backend that is managed by your internal database team. Perhaps your goal is to rip and replace the entire functionality of a legacy application and database backend with new open source technology.
|
||||
|
||||
Once you have defined a goal, you can focus your efforts. This will lead to better conversations internally as well as externally with open source database software vendors and advocates.
|
||||
|
||||
### 2. Understand your workload.
|
||||
|
||||
Despite the increasing ability of database technologies to wear many hats, each specializes in certain areas, e.g., MongoDB is now transactional, MySQL now has JSON storage. A growing trend in open source databases involves providing check boxes claiming certain features are available. One of the biggest mistakes is not using the right tool for the right job. Something leads a company down the wrong path—perhaps an overzealous developer or a manager with tunnel vision. The unfortunate thing is that the wrong tool can work fine for smaller volumes of transactions and data, but later there will be bottlenecks that can be solved only by using a different tool.
|
||||
|
||||
If you want a data analytics warehouse, an open source relational database is probably not the right choice. If you want a transaction-processing app with rigid data integrity and consistency, NoSQL options may not be the right option.
|
||||
|
||||
### 3. Don't reinvent the wheel.
|
||||
|
||||
Open source database technologies have rapidly grown, expanded, and hardened over the past several decades. We've seen a transformation from new, questionably production-ready databases to proven, enterprise-grade database backends. It's no longer necessary to be a bleeding edge, early adopter to choose open source database technologies. Organizations have grown around these communities to provide production support and tooling in the open source database space for a growing number of startups, midsized businesses, and Fortune 500 companies.
|
||||
|
||||
Battery Ventures, a tech-focused investment firm, recently introduced its [BOSS Index][2] for tracking the most popular open source projects. It's not perfect, but it provides great insight into some of the most widely adopted and active open source projects. Not surprisingly, database technologies dominate the list, comprising five of the top 10 technologies. This is a great starting point for someone new to the open source database space. A lot of times, vendors have already produced suitable architectures for solving specific problems.
|
||||
|
||||
My point is that someone has probably already done what you are trying to do. Learn from their successes and failures. Even if it is not a perfect fit, a solution can likely be modified to suit your needs. For example, Amazon provides a [CloudFormation script][3] for deploying MongoDB in its EC2 environment.
|
||||
|
||||
If you are a bleeding-edge early adopter, that doesn't mean you can't explore. If you have a unique challenge or workload that seems to fit a new open source database technology, go for it. Keep in mind that there are inherent risks (and rewards!) to being an early adopter.
|
||||
|
||||
### 4\. Start simple
|
||||
|
||||
|
||||
How many [nines][4] does your database truly need? "Achieving high availability" is often a nebulous goal for many companies. Of course, the most common answer is "it's mission-critical, and we cannot afford any downtime."
|
||||
|
||||
The more complicated your database environment, the more difficult and costly it is to manage. You can theoretically achieve higher uptime, but the tradeoffs will be the feasibility of management and performance. When in doubt, start simple. There are always options to scale out when the need arises.
|
||||
|
||||
For example, Booking.com is a widely known travel reservation site. It might be less widely known that it uses MySQL as a database backend. Nicolai Plum, a Booking.com senior systems architect, gave [a talk][5] outlining the evolution of the company's MySQL database. One of the takeaways was that the database started simple. It had to evolve over time, but in the beginning, simple master–replica architecture sufficed. As the workload and dataset increased, it introduced load balancers, multiple read replicas, archiving to Hadoop for analytics, etc. However, the early architecture was extremely simple.
|
||||
|
||||
|
||||
|
||||
### 5. When in doubt, ask an expert.
|
||||
|
||||
If you're unsure whether a database would be a good fit, reach out on forums, websites, or to vendors and strike up a conversation. This can be exciting as you research which database technologies meet your requirements and which do not. Often there are suitable alternatives that you haven't considered. The open source community is all about sharing knowledge.
|
||||
|
||||
There is one important thing to be aware of when reaching out to open source software and services vendors. Many have open-core business models that incentivize adopting their database software. Take their advice or guidance with a grain of salt and use your own ability to research, create proofs of concept, and explore alternatives.
|
||||
|
||||
### Conclusion
|
||||
|
||||
Choosing the right open source database is an important decision. Start by asking the right questions. All too often, people put the cart before the horse, making decisions before really understanding their needs.
|
||||
|
||||
Barrett Chambers will present [Choosing the Right Open Source Database][6] at [All Things Open][7], October 21-23 in Raleigh, N.C.
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/18/10/tips-choosing-right-open-source-database
|
||||
|
||||
作者：[Barrett Chambers][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/barrettc
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://www.percona.com/
|
||||
[2]: https://techcrunch.com/2017/04/07/tracking-the-explosive-growth-of-open-source-software/
|
||||
[3]: https://docs.aws.amazon.com/quickstart/latest/mongodb/welcome.html
|
||||
[4]: https://en.wikipedia.org/wiki/Five_nines
|
||||
[5]: https://www.percona.com/live/mysql-conference-2015/sessions/bookingcom-evolution-mysql-system-design
|
||||
[6]: https://allthingsopen.org/talk/choosing-the-right-open-source-database/
|
||||
[7]: https://allthingsopen.org/
|
||||
@ -0,0 +1,78 @@
|
||||
正确选择开源数据库的 5 个技巧
|
||||
======
|
||||
> 对关键应用的选择不容许丝毫错误。
|
||||
|
||||
|
||||
|
||||
你或许会遇到需要选择适合的开源数据库的情况。但这无论对于开源方面的老手或是新手，都是一项艰巨的任务。
|
||||
|
||||
在过去的几年中，采用开源技术的企业越来越多。面对这样的趋势，众多开源应用公司都纷纷承诺自己提供的解决方案能够各种问题、适应各种负载。但这些承诺不能轻信，在开源应用上的选择是重要而艰难的，尤其是数据库这种关键的应用。
|
||||
|
||||
凭借我在 [Percona][1] 和其它公司担任 IT 专家的经验，我很幸运能够指导其他人在开源技术的选择上做出正确的决策，因为需要考虑的重要因素太多了。希望通过这篇文章能够向大家分享这方面的一些技巧。
|
||||
|
||||
### 有一个明确的目标
|
||||
|
||||
这一点看似简单，但在和很多人聊过 MySQL、MongoDB、PostgreSQL 之后，我觉得这一点才是最重要的。
|
||||
|
||||
面对繁杂的开源数据库，更需要明确自己的目标。无论这个数据库是作为开发用的标准化数据库后端，抑或是用于替换遗留代码中的原有数据库，这都是一个明确的目标。
|
||||
|
||||
目标一旦确定，就可以集中精力与开源软件的提供方商讨更多细节了。
|
||||
|
||||
### 了解你的工作负载
|
||||
|
||||
尽管开源数据库技术的功能越来越丰富，但这些新加入的功能都不太具有普适性。譬如 MongoDB 新增了事务的支持、MySQL 新增了 JSON 存储的功能等等。目前开源数据库的普遍趋势是不断加入新的功能，但很多人的误区却在于没有选择最适合的工具来完成自己的工作——这样的人或许是一个自大的开发者，又或许是一个视野狭窄的主管——最终导致公司业务上的损失。最致命的是，在业务初期，使用了不适合的工具往往也可以顺利地完成任务，但随着业务的增长，很快就会到达瓶颈，尽管这个时候还可以替换更合适的工具，但成本就比较高了。
|
||||
|
||||
例如，如果你需要的是数据分析仓库，关系数据库可能不是一个适合的选择；如果你处理事务的应用要求严格的数据完整性和一致性，就不要考虑 NoSQL 了。
|
||||
|
||||
### 不要重新发明轮子
|
||||
|
||||
在过去的数十年，开源数据库技术迅速发展壮大。开源数据库从新生，到受到质疑，再到受到认可，现在已经成为很多企业生产环境的数据库。企业不再需要担心选择开源数据库技术会产生风险，因为开源数据库通常都有活跃的社区，可以为越来越多的初创公司、中型企业甚至 500 强公司提供开源数据库领域的支持和第三方工具。
|
||||
|
||||
Battery Ventures 是一家专注于技术的投资公司，最近推出了一个用于跟踪最受欢迎开源项目的 [BOSS 指数][2] 。它提供了对一些被广泛采用的开源项目和活跃的开源项目的详细情况。其中，数据库技术毫无悬念地占据了榜单的主导地位，在前十位之中占了一半。这个 BOSS 指数对于刚接触开源数据库领域的人来说，这是一个很好的切入点。当然，开源技术的提供者也会针对很多常见的典型问题给出对应的解决方案。
|
||||
|
||||
我认为，你想要做的事情很可能已经有人解决过了。即使这些先行者的解决方案不一定完全契合你的需求，但也可以从他们成功或失败案例中根据你自己的需求修改得出合适的解决方案。
|
||||
|
||||
如果你采用了一个最前沿的技术，这就是你探索的好机会了。如果你的工作负载刚好适合新的开源数据库技术，放胆去尝试吧。第一个吃螃蟹的人总是会得到意外的挑战和收获。
|
||||
|
||||
### 先从简单开始
|
||||
|
||||
你的数据库实际上需要达到多少个 [9][4] 的可用性？对许多公司来说，“实现高可用性”仅仅只是一个模糊的目标。当然，最常见的答案都会是“它是关键应用，我们无论多短的停机时间都是无法忍受的”。
|
||||
|
||||
数据库环境越复杂，管理的难度就越大，成本也会越高。理论上你总可以将数据库的可用性提得更高，但代价将会是大大增加的管理难度和性能下降。所以，先从简单开始，直到有需要时再逐步扩展。
|
||||
|
||||
例如，Booking.com 是一个有名的旅游预订网站。但少有人知的是，它使用 MySQL 作为数据库后端。 Booking.com 高级系统架构师 Nicolai Plum 曾经发表过一次[演讲][5]，讲述了他们公司使用 MySQL 数据库的历程。其中一个重点就是，在初始阶段数据库可以被配置得很简单，然后逐渐变得复杂。对于早期的数据库需求，一个简单的主从架构就足够了，但随着工作负载和数据量的增加，数据库引入了负载均衡、多个读取副本，还使用 Hadoop 进行分析。尽管如此，早期的架构仍然是非常简单的。
|
||||
|
||||
|
||||
|
||||
### 有疑问，找专家
|
||||
|
||||
如果你仍然不确定数据库选择得是否合适，可以在论坛、网站或者与软件的提供者处商讨。研究各种开源数据库是否满足自己的需求是一件很有意义的事，因为总会发现你从不知道的技术。而开源社区就是分享这些信息的地方。
|
||||
|
||||
当你接触到开源软件和软件提供者时，有一件重要的事情需要注意。很多公司都有开放的核心业务模式，鼓励采用他们的数据库软件。你可以只接受他们的部分建议和指导，然后用你自己的能力去研究和探索替代方案。
|
||||
|
||||
### 总结
|
||||
|
||||
选择正确的开源数据库是一个重要的过程。很多时候，人们都会在真正理解需求之前就做出决定，这是本末倒置的。
|
||||
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/18/10/tips-choosing-right-open-source-database
|
||||
|
||||
作者：[Barrett Chambers][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[HankChow](https://github.com/HankChow)
|
||||
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/barrettc
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://www.percona.com/
|
||||
[2]: https://techcrunch.com/2017/04/07/tracking-the-explosive-growth-of-open-source-software/
|
||||
[3]: https://docs.aws.amazon.com/quickstart/latest/mongodb/welcome.html
|
||||
[4]: https://en.wikipedia.org/wiki/Five_nines
|
||||
[5]: https://www.percona.com/live/mysql-conference-2015/sessions/bookingcom-evolution-mysql-system-design
|
||||
[6]: https://allthingsopen.org/talk/choosing-the-right-open-source-database/
|
||||
[7]: https://allthingsopen.org/
|
||||
|
||||
Loading…
Reference in New Issue
Block a user