Nvidia Shares Grace Superchip Benchmarks

By Doug Eadline

September 6, 2023

Every vendor releases benchmarks when introducing a new processor. There are lots of application-based benchmarks available for HPC, and to be sure, vendors will only release results where their product wins in the benchmarks (or mostly wins). Nvidia has released some initial benchmarks for their Grace CPU Superchip. Of course, the normal proviso applies, “take vendor benchmarks with a grain of salt,” but these initial numbers are helpful and provide some background on what we can expect from the new HPC processor.

Grace provides 72 Arm Neoverse V2 cores in a single die and is fabricated on TSMC’s 4N process node (an Nvidia-specific node). The peak performance is reported as 7.1 TFLOPs for FP64 computations. Grace is actually designed to be used in tandem with a second Grace CPU (or Hopper GPU). The two CPUs are connected over a 900 GB/s bidirectional NVLink-C2C to deliver all 144 high-performance Arm Neoverse V2 cores with up to 1 TB/s bandwidth of ECC memory. Within each Neoverse processor, Nvidia provides a Scalable Coherency Fabric (SCF) that provides a coherent and distributed cache that provides over 3.2 TB/s of total bisection bandwidth supporting local CPU cores, the NVLink-C2C, local memory, and system IO.

According to Nvidia, the Grace Superchip is in production, and OEM systems will probably be available in 2H24. This is why Nvidia released some initial benchmarks, see Figure One, to support the processor release. The benchmarks included Weather ( WRF), Molecular Dynamics (CP2K), Climate (NEMO), CFD (OpenFoam) and Graph Analytics (GapBS BFS). An AMD Genoa (2-socket Epyc 9654 with 192-cores total) and an Intel Xeon (Sapphire Rapids 8480+ with 112 cores total) were used for comparison. Note, regarding cores, the Genoa leads with 192, Superchip with 144, and Xeon with 112. On a strictly “more cores” basis, the 2-socket Genoa processors with 33% more cores should have run away with the benchmarks (memory Bandwidth not withstanding). This expectation was not the case, however. With all but one benchmark (Graph Analytics), the Genoa and Superchip maintained parity. Intel, with about 30% fewer cores, would be expected to bring up the rear.

Nvidia Grace Benchmarks
Figure One: Nvidia Grace Benchmarks (Source Nvidia)

The Graph Analytic benchmark is curious, not because the Superchip did so well, but because it does not quite fit into a mainstream HPC benchmark suite. It does show a case where the Superchip was 40% faster than the Genoa with 33% more cores. If these performance tests are independently reproduced with similar results, then the Arm Superchip outperforms the x86_64 family of processors on a core-for-core basis. The memory fabric on the Superchip probably has a lot to do with the benchmark performance.

The right-hand side of the graph is more interesting. When the processors are normalized by power consumption, the Superchip wins everything — the Graph Analytics by a whopping 250%.

We could not find any details for the 5MW Data Center projection benchmark. The “projection benchmark” phrasing is a little strange and suggests some real-world wall-outlet tests may be needed. Looking at the vendor-published power numbers does support these results, however.

  • Dual Nvidia Grace Superchip: 500W
  • Dual AMD EPYC 9654: 720W
  • Dual Intel Xeon 8480+: 700W

Note: Power ratings for most processors are reported as TDP (Thermal Design Power), but each vendor defines TDP differently, and the actual electrical power use can often be much higher. A good rule of thumb is the actual electrical power draw (in Watts) can be as high as 1.5 times the stated TDP wattage specification.

For now, we will consider the TDP measurements roughly equivalent and give Nvidia credit for a power-efficient processor (compared to x86_64 options). For a complete measure of power, the entire server needs to be taken into account, which is why the best measurement is the Kill-O-Watt  wall outlet test.

If the benchmarks hold up and are supplemented with other independent tests, the Nvidia Grace CPU should be a serious contender for high throughput computing. The still undetermined aspect of the Superchip will be price. The commodity market does provide an advantage to the x86_64 market. Still, the dual-purpose architecture of the Grace CPU (either coupled with another Grace CPU or with a Hopper GPU) may make the economical work for large customers.

Finally, and probably not visible in Figure One above, the copyright is listed as Arm, not Nvidia. 


Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

AI Saves the Planet this Earth Day

April 22, 2024

Earth Day was originally conceived as a day of reflection. Our planet’s life-sustaining properties are unlike any other celestial body that we’ve observed, and this day of contemplation is meant to provide all of us Read more…

Intel Announces Hala Point – World’s Largest Neuromorphic System for Sustainable AI

April 22, 2024

As we find ourselves on the brink of a technological revolution, the need for efficient and sustainable computing solutions has never been more critical.  A computer system that can mimic the way humans process and s Read more…

Empowering High-Performance Computing for Artificial Intelligence

April 19, 2024

Artificial intelligence (AI) presents some of the most challenging demands in information technology, especially concerning computing power and data movement. As a result of these challenges, high-performance computing Read more…

Kathy Yelick on Post-Exascale Challenges

April 18, 2024

With the exascale era underway, the HPC community is already turning its attention to zettascale computing, the next of the 1,000-fold performance leaps that have occurred about once a decade. With this in mind, the ISC Read more…

2024 Winter Classic: Texas Two Step

April 18, 2024

Texas Tech University. Their middle name is ‘tech’, so it’s no surprise that they’ve been fielding not one, but two teams in the last three Winter Classic cluster competitions. Their teams, dubbed Matador and Red Read more…

2024 Winter Classic: The Return of Team Fayetteville

April 18, 2024

Hailing from Fayetteville, NC, Fayetteville State University stayed under the radar in their first Winter Classic competition in 2022. Solid students for sure, but not a lot of HPC experience. All good. They didn’t Read more…

AI Saves the Planet this Earth Day

April 22, 2024

Earth Day was originally conceived as a day of reflection. Our planet’s life-sustaining properties are unlike any other celestial body that we’ve observed, Read more…

Kathy Yelick on Post-Exascale Challenges

April 18, 2024

With the exascale era underway, the HPC community is already turning its attention to zettascale computing, the next of the 1,000-fold performance leaps that ha Read more…

Software Specialist Horizon Quantum to Build First-of-a-Kind Hardware Testbed

April 18, 2024

Horizon Quantum Computing, a Singapore-based quantum software start-up, announced today it would build its own testbed of quantum computers, starting with use o Read more…

MLCommons Launches New AI Safety Benchmark Initiative

April 16, 2024

MLCommons, organizer of the popular MLPerf benchmarking exercises (training and inference), is starting a new effort to benchmark AI Safety, one of the most pre Read more…

Exciting Updates From Stanford HAI’s Seventh Annual AI Index Report

April 15, 2024

As the AI revolution marches on, it is vital to continually reassess how this technology is reshaping our world. To that end, researchers at Stanford’s Instit Read more…

Intel’s Vision Advantage: Chips Are Available Off-the-Shelf

April 11, 2024

The chip market is facing a crisis: chip development is now concentrated in the hands of the few. A confluence of events this week reminded us how few chips Read more…

The VC View: Quantonation’s Deep Dive into Funding Quantum Start-ups

April 11, 2024

Yesterday Quantonation — which promotes itself as a one-of-a-kind venture capital (VC) company specializing in quantum science and deep physics  — announce Read more…

Nvidia’s GTC Is the New Intel IDF

April 9, 2024

After many years, Nvidia's GPU Technology Conference (GTC) was back in person and has become the conference for those who care about semiconductors and AI. I Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…


How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Leading Solution Providers


Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Eyes on the Quantum Prize – D-Wave Says its Time is Now

January 30, 2024

Early quantum computing pioneer D-Wave again asserted – that at least for D-Wave – the commercial quantum era has begun. Speaking at its first in-person Ana Read more…

GenAI Having Major Impact on Data Culture, Survey Says

February 21, 2024

While 2023 was the year of GenAI, the adoption rates for GenAI did not match expectations. Most organizations are continuing to invest in GenAI but are yet to Read more…

The GenAI Datacenter Squeeze Is Here

February 1, 2024

The immediate effect of the GenAI GPU Squeeze was to reduce availability, either direct purchase or cloud access, increase cost, and push demand through the roof. A secondary issue has been developing over the last several years. Even though your organization secured several racks... Read more…

Intel’s Xeon General Manager Talks about Server Chips 

January 2, 2024

Intel is talking data-center growth and is done digging graves for its dead enterprise products, including GPUs, storage, and networking products, which fell to Read more…

  • arrow
  • Click Here for More Headlines
  • arrow