As part of the company's Supercomputing 2018, a new FPGA accelerator card was announced by Xilinx. The Xilinx Alveo U280 is one of the company's pre-ACAP 16nm UltraScale+ architecture FPGA products. The U280 features 8GB of Samsung High Bandwidth Memory (HBM2) plus 32GB of DDR4 memory. The goal of the new card is to accelerate database search and analytics, machine learning inference, and other memory-bound applications.
Buried in the documentation for the card is a nugget of extremely interesting information:
"The U280 acceleration card includes CCIX support to leverage existing server interconnect infrastructure for high bandwidth, low latency cache coherent shared memory access with CCIX enabled processors including Arm and AMD." (Source: Xilinx Alveo U280 whitepaper WP50 (v1.0) accessed 16 November 2018)
We were recently at the AMD Next Horizon Event and STH friend Dr. Ian Dr. Ian Cutress at Anandtech (not a typo, that is what his SC18 badge said) touched upon this in his interview with AMD CTO Mark Papermaster. Neither in the Rome disclosure nor the interview did AMD confirm CCIX support. However, AMD publicly supports CCIX and Gen-Z and when we asked if this means Rome supports CCIX all we received was that AMD supports CCIX but has not announced a product with it yet. Arm may have chips derived from its IP with CCIX support, but AMD has a more well-defined roadmap.
https://www.servethehome.com/xilinx-alveo-u280-launched-possibly-with-amd-epyc-ccix-support/
(Score: 0) by Anonymous Coward on Wednesday November 21 2018, @08:24PM (1 child)
https://en.wikichip.org/wiki/ccix [wikichip.org]
(Score: 0) by Anonymous Coward on Thursday November 22 2018, @02:59PM
thanks for editing the web to point to wikipedia ^_^
(Score: 0) by Anonymous Coward on Wednesday November 21 2018, @08:50PM
So it sounds like the plan is to replace Intel CPU + Nvidia GPU with AMD CPU + Xlinix FPGA in the machine learning server space. I wonder what this means for AMD's open source answer to CUDA, ROCm? Do they still plan on competing GPU-wise with Nvidia?
(Score: 1) by schusselig on Thursday November 22 2018, @08:08AM (1 child)
> What-*does*-FPGA-stand-for? dept.
Field Programmable Gate Array [wikipedia.org] which is basically like an ASIC that can be fiddled with after manufacturing.
(Score: 0) by Anonymous Coward on Thursday November 22 2018, @03:02PM
i suppose that's what is needed for the regular A.I. to gain consciousness?