AMD Instinct MI200 CDNA 2 MCM GPU Is a Beast: 1.7 GHz Clocks, 47.9 TFLOPs FP64

Flint 1760
  • Total Posts : 5804
  • Reward points : 0
  • Joined: 2009/04/26 15:44:26
  • Status: offline
  • Ribbons : 22
2021/10/24 05:40:34 (permalink)
wccftech -  AMD Instinct MI200 CDNA 2 MCM GPU Is a Beast: 1.7 GHz Clocks, 47.9 TFLOPs FP64 & Over 4X Increase In FP64/BF16 Performance Over MI100
“AMD's flagship Instinct MI200 is on the verge of launch and it will be the first GPU for the HPC segment to feature an MCM design based on the CDNA 2 architecture. It looks like the GPU will offer some insane performance numbers compared to the existing Instinct MI100 GPU with a 4x increase in FP16 compute.
It looks like the Instinct MI200 lineup will include two variants, a standard MI250, and a MI250X. According to the details, the MI250X will get 110 CUs per die (220 CUs in total), 128 GB HBM2e memory, a 500W TDP and will be based on 7nm.
The AMD Instinct MI200 will be have a clock speed of up to 1.7 GHz which is a 13% increase over the Instinct MI100. The CDNA 2 powered MCM GPU also has almost twice the number of stream processors at 14,080 cores, packed within 220 Compute Units. While it was expected that the GPU would have 240 Compute units with 15,360 cores, the config is replaced by a cut-down variant due to yields. With that said, it is possible that we may see the full SKU launch in the future, offering even higher performance.
In terms of performance, the AMD Instinct MI200 HPC Accelerator is going to offer almost 50 TFLOPs (47.9) TFLOPs of FP64 & FP32 compute horsepower. Versus the Instinct MI100, this is a 4.16x increase in the FP64 segment. In fact, the FP64 numbers of the MI200 exceed the FP32 performance of its predecessor. Moving over to the FP16 and BF16 numbers, we are looking at an insane 383 TFLOPs of performance. For perspective, the MI100 only offers 92.3 TFLOPs of peak BFloat16 performance and 184.6 TFLOPs peak FP16 performance.
The AMD Instinct MI200 will be powering three top-tier supercomputers which include the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. The competition includes the A100 80 GB which offers 19.5 TFLOPs of FP64, 156 TFLOPs of FP32 and 312 TFLOPs of FP16 compute power. But we are likely to hear about NVIDIA's own Hopper MCM GPU next year so there's going to be a heated competition between the two GPU juggernauts in 2022.”

My thoughts: The HPC market revenue in 2021 is predicted to be $26.77B U.S. and by 2024, HPC market revenues are forecast to rise to $38.21B U.S.  This market is expanding and as another example of the importance of the HPC market, TSMC’s revenue from HPC for 2Q21 was 39% of their total, which is a 12% increase over the quarter.


0 Replies Related Threads

    Jump to:
  • Back to Mobile