It appears you have not yet registered with our community. To register please click here...

 
Go Back [M] > Madshrimps > WebNews
IBM AC922 Power 9 server has 6 Nvidia V100s IBM AC922 Power 9 server has 6 Nvidia V100s
FAQ Members List Calendar Search Today's Posts Mark Forums Read


IBM AC922 Power 9 server has 6 Nvidia V100s
Reply
 
Thread Tools
Old 10th May 2018, 13:36   #1
[M] Reviewer
 
Stefan Mileschin's Avatar
 
Join Date: May 2010
Location: Romania
Posts: 148,055
Stefan Mileschin Freshly Registered
Default IBM AC922 Power 9 server has 6 Nvidia V100s

Enterprise AI beats the X86 competition

IBM has just announced an AC922 Power 9 based machine that addresses artificial intelligence (AI), deep and machine learning - aimed at the enterprise AI market.

The new AC922 server has four to six V100 GPUs with 16 or 32GB and comes with a new Power 9 processor, supports PCIe Gen 4 and the second generation NVlink 2.0 interlink unique to IBM’s platform. The system is targeting the memory coherence as the I/O system gets you to use the 2TB of system RAM almost as fast as the access time of the GPU RAM.

IBM’s four GPU system enables an incredible CPU to GPU bandwidth of 150 GB/s/ CPU to DDR4 memory can only achieve 170 GB/s, so NVlink 2.0 is actually very close to that speed. NVlink 2.0 lets two V100 cards talk to the same 150 GB/s. This is the unique ability of the IBM Power 9 platform.
2TB plus Hash tables data sets

This gives great coherent access to system memory. The whole system address 2TB and this should be enough for very large hash tables. Of course, the system comes with PCIe Gen 4 CAPI 2.0 support. The CAPI 2.0 - unique to the Power 9 processor - enables 32GB /s speeds. CAPI 2.0 Coherent Accelerator Processor Interface is designed as a layer on top of the PCIe that enables direct connection of the CPUI and external accelerations like GPU, ASIC, FPGA or fast storage.

The AC922 accelerated server machine also supports an open source variant of CAPI. OpenCAPI is an Open Standard-based High-speed interconnect for accelerators, storage-class memory, and networking.

The key advantage over the X86 competition - read Intel) -is the memory coherency as you can directly attach the GPU to the system memory. IBM’s advantage is that it can interconnect the GPU to system memory much faster than Intel based machines. The four GPU machine is equipped with InfiniBand and Ethernet and comes as an air-cooled solution.

https://fudzilla.com/news/ai/46244-i...-6-nvidia-v100
Stefan Mileschin is offline   Reply With Quote
Reply


Similar Threads
Thread Thread Starter Forum Replies Last Post
Nvidia announces DGX Station with four Volta V100s Stefan Mileschin WebNews 0 12th May 2017 10:42
NVIDIA Unveils the DGX-1 HPC Server: 8 Teslas, 3U, Q2 2016 Stefan Mileschin WebNews 0 7th April 2016 12:47
Intel releases low power server chips Stefan Mileschin WebNews 0 24th July 2013 07:32
Intel releases low power server chips Stefan Mileschin WebNews 0 24th July 2013 07:26
DIY Power Supply for Home Server jmke WebNews 0 28th March 2011 15:32
AMD Beats Intel in 45nm Low Power Server Test jmke WebNews 0 20th March 2009 22:13
Windows Home Server Power Back 1 Released jmke WebNews 1 22nd July 2008 14:24
Asustek Launches Power Efficient Server Board Shogun WebNews 0 26th December 2007 09:17
Build a Low-Cost, Low-Power Home Server jmke WebNews 0 14th December 2006 12:54
Group to gauge server power consumption Sidney WebNews 0 1st April 2006 19:32

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


All times are GMT +1. The time now is 07:04.


Powered by vBulletin® - Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO