mirror of
https://github.com/LCTT/TranslateProject.git
synced 2024-12-26 21:30:55 +08:00
3413267d35
sources/talk/20190812 Xilinx launches new FPGA cards that can match GPU performance.md
70 lines
4.9 KiB
Markdown
70 lines
4.9 KiB
Markdown
[#]: collector: (lujun9972)
|
||
[#]: translator: ( )
|
||
[#]: reviewer: ( )
|
||
[#]: publisher: ( )
|
||
[#]: url: ( )
|
||
[#]: subject: (Xilinx launches new FPGA cards that can match GPU performance)
|
||
[#]: via: (https://www.networkworld.com/article/3430763/xilinx-launches-new-fpga-cards-that-can-match-gpu-performance.html)
|
||
[#]: author: (Andy Patrizio https://www.networkworld.com/author/Andy-Patrizio/)
|
||
|
||
Xilinx launches new FPGA cards that can match GPU performance
|
||
======
|
||
Xilinx says its new FPGA card, the Alveo U50, can match the performance of a GPU in areas of artificial intelligence (AI) and machine learning.
|
||
![Thinkstock][1]
|
||
|
||
Xilinx has launched a new FPGA card, the Alveo U50, that it claims can match the performance of a GPU in areas of artificial intelligence (AI) and machine learning.
|
||
|
||
The company claims the card is the industry’s first low-profile adaptable accelerator with PCIe Gen 4 support, which offers double the throughput over PCIe Gen3. It was finalized in 2017, but cards and motherboards to support it have been slow to come to market.
|
||
|
||
The Alveo U50 provides customers with a programmable low-profile and low-power accelerator platform built for scale-out architectures and domain-specific acceleration of any server deployment, on premises, in the cloud, and at the edge.
|
||
|
||
**[ Also read: [What is quantum computing (and why enterprises should care)][2] ]**
|
||
|
||
Xilinx claims the Alveo U50 delivers 10 to 20 times improvements in throughput and latency as compared to a CPU. One thing's for sure, it beats the competition on power draw. It has a 75 watt power envelope, which is comparable to a desktop CPU and vastly better than a Xeon or GPU.
|
||
|
||
For accelerated networking and storage workloads, the U50 card helps developers identify and eliminate latency and data movement bottlenecks by moving compute closer to the data.
|
||
|
||
![Xilinx Alveo U50][3]
|
||
|
||
The Alveo U50 card is the first in the Alveo portfolio to be packaged in a half-height, half-length form factor. It runs the Xilinx UltraScale+ FPGA architecture, features high-bandwidth memory (HBM2), 100 gigabits per second (100 Gbps) networking connectivity, and support for the PCIe Gen 4 and CCIX interconnects. Thanks to the 8GB of HBM2 memory, data transfer speeds can reach 400Gbps. It also supports NVMe-over-Fabric for high-speed SSD transfers.
|
||
|
||
That’s a lot of performance packed into a small card.
|
||
|
||
**[ [Get certified as an Apple Technical Coordinator with this seven-part online course from PluralSight.][4] ]**
|
||
|
||
### What the Xilinx Alveo U50 can do
|
||
|
||
Xilinx is making some big boasts about Alveo U50's capabilities:
|
||
|
||
* Deep learning inference acceleration (speech translation): delivers up to 25x lower latency, 10x higher throughput, and significantly improved power efficiency per node compared to GPU-only for speech translation performance.
|
||
* Data analytics acceleration (database query): running the TPC-H Query benchmark, Alveo U50 delivers 4x higher throughput per hour and reduced operational costs by 3x compared to in-memory CPU.
|
||
* Computational storage acceleration (compression): delivers 20x more compression/decompression throughput, faster Hadoop and big data analytics, and over 30% lower cost per node compared to CPU-only nodes.
|
||
* Network acceleration (electronic trading): delivers 20x lower latency and sub-500ns trading time compared to CPU-only latency of 10us.
|
||
* Financial modeling (grid computing): running the Monte Carlo simulation, Alveo U50 delivers 7x greater power efficiency compared to GPU-only performance for a faster time to insight, deterministic latency and reduced operational costs.
|
||
|
||
|
||
|
||
The Alveo U50 is sampling now with OEM system qualifications in process. General availability is slated for fall 2019.
|
||
|
||
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/3430763/xilinx-launches-new-fpga-cards-that-can-match-gpu-performance.html
|
||
|
||
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/2014/04/bolts-of-light-speeding-through-the-acceleration-tunnel-95535268-100264665-large.jpg
|
||
[2]: https://www.networkworld.com/article/3275367/what-s-quantum-computing-and-why-enterprises-need-to-care.html
|
||
[3]: https://images.idgesg.net/images/article/2019/08/xilinx-alveo-u50-100808003-medium.jpg
|
||
[4]: https://pluralsight.pxf.io/c/321564/424552/7490?u=https%3A%2F%2Fwww.pluralsight.com%2Fpaths%2Fapple-certified-technical-trainer-10-11
|
||
[5]: https://www.facebook.com/NetworkWorld/
|
||
[6]: https://www.linkedin.com/company/network-world
|