# AMD GPU and CPU top 1 supercomputer, 1.1 exaflops (and other 10+ "AI exaflops")

**URL:** https://discourse.julialang.org/t/amd-gpu-and-cpu-top-1-supercomputer-1-1-exaflops-and-other-10-ai-exaflops/83527
**Category:** Offtopic
**Created:** [June 29, 2022, 4:54pm UTC](https://discourse.julialang.org/t/amd-gpu-and-cpu-top-1-supercomputer-1-1-exaflops-and-other-10-ai-exaflops/83527 "2022-06-29T16:54:45Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![Palli](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/palli/32/3380_2.png) [@Palli](https://discourse.julialang.org/u/Palli)
#### Post date: [June 29, 2022, 4:54pm UTC](https://discourse.julialang.org/t/amd-gpu-and-cpu-top-1-supercomputer-1-1-exaflops-and-other-10-ai-exaflops/83527/1 "2022-06-29T16:54:45Z")

</div>

In case people missed it, there’s a new top 1, exascale, supercomputer out, Frontier.

A.  
And it’s 50% more power-efficient than anything out there. Intel and Nvidia are almost gone from the top 10 list of top500 and green500 (AMD has the top 4 computers there).

AMD can seemingly take all the credit, since both the GPUs and CPUs are AMD based. However other things factor in too, e.g. the interconnect: Slingshot 11, which I hadn’t heard about.

> [@CUDA.jl - Better GPU but Worse Performance](https://discourse.julialang.org/t/cuda-jl-better-gpu-but-worse-performance/83337/7):
>
> +1 to @LaurentPlagne Gaming GPUs are best for FP16 and FP32. They are not performant for FP64 and are not designed for simulation workloads per se. For FP64 you’d want to explore GPUs like Tesla V100 and A100s. P2000 would be quite slow as well in comparison if you have large simulation workload.

> For FP64 you’d want to explore GPUs like Tesla V100 and A100s.

Well, I would explore AMD… and also Metal.jl. Anyone know what’s responsible for the 50%+ more efficiency (and much faster Float64)? I don’t think it can be the interconnect (unless others really hold performance back, and all are limited by the same speed of light).

Note, HP which makes the AMD-based supercomputers, bought Cray in 2019 if I recall, and it’s their interconnect (so deserve some credit too):

> **[HPE Slingshot Interconnect – Your HPC Networking Solution](https://www.hpe.com/us/en/compute/hpc/slingshot-interconnect.html)**
>
> HPE Slingshot Interconnect is your networking solution for your high performance computing environment.

> A high performance network for HPE Cray supercomputers and HPE HPC clusters designed for exascale era supercomputing […]  
> Ethernet compatible, HPE Slingshot interconnect enables straightforward execution of cloudlike and converged workloads in a [**high performance computing**](https://www.hpe.com/us/en/what-is/high-performance-computing.html) environment.

B.  
It’s unclear to me if the already “existing” computer has “20 AI exaflops”, or bad reporting.

> **[Taiwan’s Tech Titans Adopt World’s First NVIDIA Grace CPU-Powered System Designs](https://nvidianews.nvidia.com/news/taiwans-tech-titans-adopt-worlds-first-nvidia-grace-cpu-powered-system-designs)**
>
> NVIDIA today announced that Taiwan’s leading computer makers are set to release the first wave of systems powered by the NVIDIA Grace™ CPU Superchip and Grace Hopper Superchip for a wide range of workloads spanning digital twins, AI, high performance...

> The Grace CPU Superchip features two CPU chips, connected coherently through an NVIDIA [NVLink®-C2C interconnect](https://www.nvidia.com/en-us/data-center/nvlink-c2c/), with up to 144 high-performance Arm V9 cores with scalable vector extensions and a 1 terabyte-per-second memory subsystem.

> **[Nvidia's Grace CPU Superchip to Power Two Supercomputers, Up to Ten 'AI...](https://www.tomshardware.com/news/nvidias-grace-cpu-superchip-to-power-two-supercomputers-up-to-ten-ai-exaflops)**
>
> Arm-based Grace CPUs and Hopper GPUs

Strange statement: “The Swiss National Computing Center’s existing [20 AI ExaFlops Alps supercomputer](https://click.linksynergy.com/deeplink?id=kXQk6%2AivFEQ&mid=44270&u1=tomshardware-row-1146975763456496500&murl=https%3A%2F%2Fnvidianews.nvidia.com%2Fnews%2Fswiss-national-supercomputing-centre-hewlett-packard-enterprise-and-nvidia-announce-worlds-most-powerful-ai-capable-supercomputer) will”

> **[Swiss National Supercomputing Centre, Hewlett Packard Enterprise and NVIDIA...](https://nvidianews.nvidia.com/news/swiss-national-supercomputing-centre-hewlett-packard-enterprise-and-nvidia-announce-worlds-most-powerful-ai-capable-supercomputer)**
>
> The Swiss National Supercomputing Centre (CSCS), Hewlett Packard Enterprise (HPE) and NVIDIA today announced that they are creating what is expected to be...

> Taking advantage of the tight coupling between NVIDIA CPUs and GPUs, Alps will be able to train GPT-3, one of the world’s largest natural language processing models, in only two days — 7x faster than NVIDIA’s 2.8-AI exaflops Selene supercomputer, currently recognized as the world’s leading supercomputer for AI by MLPerf.

It’s unclear to me if the new ARM-based **CPUs** are based on AWS’s since previous announcement: “NVIDIA GPU + AWS Graviton2-Based Amazon”.