Nvidia Serves Up Its First Arm Datacenter CPU ‘Grace’ During Kitchen Keynote

By Tiffany Trader

April 12, 2021

Today at Nvidia’s annual spring GPU Technology Conference (GTC), held virtually once more due to the pandemic, the company unveiled its first ever Arm-based CPU, called Grace in honor of the famous American programmer Grace Hopper. The announcement of the new Arm CPU follows Nvidia’s 2019 declaration of intent to fully embrace Arm and its September 2020 bid to acquire Arm for $40 billion.

Grace is expected to debut in 2023 with two HPC centers leading the way. The Swiss National Supercomputing Centre (CSCS) and the U.S. Department of Energy’s Los Alamos National Laboratory are the first to announce plans to build Grace-powered supercomputers in partnership with HPE and Nvidia.

Using future-generation Arm Neoverse cores and next-generation Nvidia NVLink interconnect technology, Grace has been designed for tight coupling with Nvidia GPUs to power the very largest AI and HPC workloads, according to Nvidia.

In his third virtual GTC “kitchen keynote,” Nvidia CEO Jensen Huang said the chip, combined with Nvidia’s GPUs and high-performance networking from its Mellanox division, gives Nvidia “the third foundational technology for computing, and the ability to re-architect every aspect of the datacenter for AI.”

Using fourth-generation NVLink technology, Grace enables 900 GB/s of bidirectional bandwidth between the CPU and GPU, driving significantly higher aggregate bandwidth over today’s standard servers (~30x higher says Nvidia). The new architecture also provides cache coherence with a single memory address space, unifying system and HBM GPU memory to simplify programmability.

“Grace highlights the beauty of Arm,” Huang said. “Their IP model allowed us to create the optimal CPU for this application, which achieves x-factor speed up.” He said the Grace CPU will deliver a score of over 300 on the SPECrate2017_int_base benchmark and over 2,400 SPECrate2017_int_base CPU performance per eight-GPU DGX. In comparison, today’s eight-GPU DGX A100 achieves 450 SPECint rate.

Unlike most GPU-accelerated systems on the market today, which have a two-to-one or higher ratio of GPUs to CPUs (with four-to-one being something of a sweet spot), Grace-based systems will be architected with a one-to-one ratio of CPU to GPU. While the company is not yet announcing products, based on Jensen’s SPECint performance claims, it seems an eight-GPU DGX server — with eight Grace CPUs — is in the works.

Nvidia’s fourth-generation NVLink fabric connects CPU to CPU, CPU to GPU and GPU to GPU. The only other CPU to offer native NVLink support is the IBM Power platform (Power8+ and Power9). IBM’s NVLink’d Power9 server (AC922) forms the basis of the Summit and Sierra supercomputers (currently ranked #2 and #3 in the world), installed at Oak Ridge National Lab and Lawrence Livermore Lab, respectively.

Grace also has a new memory subsystem, leveraging LPDDR5X memory technology, which has has twice the bandwidth of today’s DDR4, and is 10 times more energy efficient, according to Nvidia. “We optimize this memory subsystem to support server class reliability through mechanisms like ECC and redundancy,” said Paresh Kharya, senior director of accelerated computing at Nvidia, in a pre-briefing last week.

“This efficiency means you can divert more power towards compute rather than moving the bits around,” said Kharya.

Grace will be supported by Nvidia’s HPC software development kit and its CUDA and CUDA-X libraries.

CSCS and Los Alamos both have Grace-based supercomputers under development with expected delivery in 2023. The CSCS “Alps” system is being billed as the world’s most powerful AI-capable supercomputer, expected to deliver 20 exaflops of performance for AI, using Nvidia’s mixed-precision arithmetic and sparsity features. Based on the HPE Cray XE (formerly Shasta) architecture, Alps will advance the boundaries of whole-earth scale weather and climate simulation, quantum chemistry and quantum physics for the Large Hadron Collider.

Nvidia reports that due to its scale and tight coupling between the CPUs and GPUs, Alps will be able to train the massive GPT-3 language processing model in only two days. That is seven times faster than the Nvidia Selene supercomputer, which is currently ranked number five on the Top500 (with 63.5 Linpack petaflops and 2.8 “AI exaflops”), according to Nvidia.

(Read about CSCS’s software-defined strategy for Alps in this interview with the center’s director Thomas Schultess.)

Scientists at Los Alamos report they are taking delivery of Nvidia A100 GPUs as a first step to receiving a Grace CPU-based system that will facilitate modeling, simulation, and data analysis in support of the lab’s mission. Los Alamos expects to be the first U.S. customer for the new Grace CPUs and will be part of a multi-year codesign collaboration that will inform hardware and software design choices for the benefit of scientific discovery. The lab’s Grace system is also being built by HPE, implementing its Cray EX architecture.

“We’re thrilled by the enthusiasm of the supercomputing community, welcoming us to make Arm a top-notch scientific computing platform,” said Huang today.

“Arm is the most popular CPU in the world, for good reason. It’s super energy-efficient and it’s open licensing model inspires a world of innovators to create product around it,” the CEO said.

Nvidia’s roadmap now includes three chips: the GPU, CPU and DPU. “Each chip architecture has a two-year rhythm, and likely a kicker in between,” Huang said. “One year we’ll focus on x86 platforms, one year we’ll focus on Arm platforms. The Nvidia architecture and platforms will support x86 and Arm, whatever customers and markets prefer.”

The arrival of Grace has in some sense been a decade in the making, stretching back to Nvidia’s 2011 “Project Denver,” the company’s plan to build an integrated CPU+GPU processor (with Arm Neoverse and Nvidia GPU cores) capable of powering personal computers, workstations, servers and supercomputers. The full scope of that project wasn’t realized, but Nvidia did end up making Arm+GPU chips (Tegra/Xaviar and Jetson), for the embedded worlds of mobile, robotics, portable gaming and autonomous vehicles.

In addition to revealing its very own Arm CPU today, Nvidia continues to strengthen its support of Arm-based technologies with partners. Huang announced that together with Amazon Web Services, it is bringing Graviton2 Arm CPUs and Nvidia GPUs together in an EC2 instance, expected later this year. The new instances target demanding cloud workloads, AI, and cloud gaming, said Huang.

Nvidia also announced a partnership with Ampere Computing to create a scientific and cloud computing SDK and reference system. Ampere Computing’s Altra CPU has 80 Neoverse-N1 cores and delivers 285 SPECint rate, “right up there with the highest performance x86,” said Huang.

In addition, Nvidia said it’s entered into a partnership with chip company Marvell to create an edge and enterprise computing SDK and reference system. Marvell’s Octeon chip targets IO storage and 5G processing, and the system is ideal for hyperconverged edge servers, noted Huang.

Absent from today’s news raft was the Cambridge AI research center announced last September, the centerpiece of which is to be an Arm-based supercomputer. Nvidia told HPCwire that the project is still on track, but did not disclose any further details. In October 2020, a company representative told HPCwire “plans are still evolving for the [Cambridge] Arm-based supercomputer,” and said the project was not tied to the closing of the Arm acquisition.

Nvidia also shared that the Cambridge-1 AI SuperPod computer is approaching readiness with updates likely to made during the GTC21 conference proceedings.

With robust support for Arm across its entire ecosystem and the debut of a homegrown Arm CPU, all of the pieces are falling into place for Nvidia’s full-stack datacenter solution. While the pending deal to acquire Arm is still under review, Nvidia is showing it has a strong Arm play with or without actually owning Arm — and that after all is the beauty of the IP licensing model.

Subscribe to HPCwire's Weekly Update!

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

Why HPC Storage Matters More Now Than Ever: Analyst Q&A

September 17, 2021

With soaring data volumes and insatiable computing driving nearly every facet of economic, social and scientific progress, data storage is seizing the spotlight. Hyperion Research analyst and noted storage expert Mark No Read more…

GigaIO Gets $14.7M in Series B Funding to Expand Its Composable Fabric Technology to Customers

September 16, 2021

Just before the COVID-19 pandemic began in March 2020, GigaIO introduced its Universal Composable Fabric technology, which allows enterprises to bring together any HPC and AI resources and integrate them with networking, Read more…

What’s New in HPC Research: Solar Power, ExaWorks, Optane & More

September 16, 2021

In this regular feature, HPCwire highlights newly published research in the high-performance computing community and related domains. From parallel programming to exascale to quantum computing, the details are here. Read more…

Cerebras Brings Its Wafer-Scale Engine AI System to the Cloud

September 16, 2021

Five months ago, when Cerebras Systems debuted its second-generation wafer-scale silicon system (CS-2), co-founder and CEO Andrew Feldman hinted of the company’s coming cloud plans, and now those plans have come to fruition. Today, Cerebras and Cirrascale Cloud Services are launching... Read more…

AI Hardware Summit: Panel on Memory Looks Forward

September 15, 2021

What will system memory look like in five years? Good question. While Monday's panel, Designing AI Super-Chips at the Speed of Memory, at the AI Hardware Summit, tackled several topics, the panelists also took a brief glimpse into the future. Unlike compute, storage and networking, which... Read more…

AWS Solution Channel

Supporting Climate Model Simulations to Accelerate Climate Science

The Amazon Sustainability Data Initiative (ASDI), AWS is donating cloud resources, technical support, and access to scalable infrastructure and fast networking providing high performance computing (HPC) solutions to support simulations of near-term climate using the National Center for Atmospheric Research (NCAR) Community Earth System Model Version 2 (CESM2) and its Whole Atmosphere Community Climate Model (WACCM). Read more…

ECMWF Opens Bologna Datacenter in Preparation for Atos Supercomputer

September 14, 2021

In January 2020, the European Centre for Medium-Range Weather Forecasts (ECMWF) – a juggernaut in the weather forecasting scene – signed a four-year, $89-million contract with European tech firm Atos to quintuple its supercomputing capacity. With the deal approaching the two-year mark, ECMWF... Read more…

Why HPC Storage Matters More Now Than Ever: Analyst Q&A

September 17, 2021

With soaring data volumes and insatiable computing driving nearly every facet of economic, social and scientific progress, data storage is seizing the spotlight Read more…

Cerebras Brings Its Wafer-Scale Engine AI System to the Cloud

September 16, 2021

Five months ago, when Cerebras Systems debuted its second-generation wafer-scale silicon system (CS-2), co-founder and CEO Andrew Feldman hinted of the company’s coming cloud plans, and now those plans have come to fruition. Today, Cerebras and Cirrascale Cloud Services are launching... Read more…

AI Hardware Summit: Panel on Memory Looks Forward

September 15, 2021

What will system memory look like in five years? Good question. While Monday's panel, Designing AI Super-Chips at the Speed of Memory, at the AI Hardware Summit, tackled several topics, the panelists also took a brief glimpse into the future. Unlike compute, storage and networking, which... Read more…

ECMWF Opens Bologna Datacenter in Preparation for Atos Supercomputer

September 14, 2021

In January 2020, the European Centre for Medium-Range Weather Forecasts (ECMWF) – a juggernaut in the weather forecasting scene – signed a four-year, $89-million contract with European tech firm Atos to quintuple its supercomputing capacity. With the deal approaching the two-year mark, ECMWF... Read more…

Quantum Computer Market Headed to $830M in 2024

September 13, 2021

What is one to make of the quantum computing market? Energized (lots of funding) but still chaotic and advancing in unpredictable ways (e.g. competing qubit tec Read more…

Amazon, NCAR, SilverLining Team for Unprecedented Cloud Climate Simulations

September 10, 2021

Earth’s climate is, to put it mildly, not in a good place. In the wake of a damning report from the Intergovernmental Panel on Climate Change (IPCC), scientis Read more…

After Roadblocks and Renewals, EuroHPC Targets a Bigger, Quantum Future

September 9, 2021

The EuroHPC Joint Undertaking (JU) was formalized in 2018, beginning a new era of European supercomputing that began to bear fruit this year with the launch of several of the first EuroHPC systems. The undertaking, however, has not been without its speed bumps, and the Union faces an uphill... Read more…

How Argonne Is Preparing for Exascale in 2022

September 8, 2021

Additional details came to light on Argonne National Laboratory’s preparation for the 2022 Aurora exascale-class supercomputer, during the HPC User Forum, held virtually this week on account of pandemic. Exascale Computing Project director Doug Kothe reviewed some of the 'early exascale hardware' at Argonne, Oak Ridge and NERSC (Perlmutter), while Ti Leggett, Deputy Project Director & Deputy Director... Read more…

Ahead of ‘Dojo,’ Tesla Reveals Its Massive Precursor Supercomputer

June 22, 2021

In spring 2019, Tesla made cryptic reference to a project called Dojo, a “super-powerful training computer” for video data processing. Then, in summer 2020, Tesla CEO Elon Musk tweeted: “Tesla is developing a [neural network] training computer called Dojo to process truly vast amounts of video data. It’s a beast! … A truly useful exaflop at de facto FP32.” Read more…

Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

May 27, 2021

A ribbon-cutting ceremony held virtually at Berkeley Lab's National Energy Research Scientific Computing Center (NERSC) today marked the official launch of Perlmutter – aka NERSC-9 – the GPU-accelerated supercomputer built by HPE in partnership with Nvidia and AMD. Read more…

Esperanto, Silicon in Hand, Champions the Efficiency of Its 1,092-Core RISC-V Chip

August 27, 2021

Esperanto Technologies made waves last December when it announced ET-SoC-1, a new RISC-V-based chip aimed at machine learning that packed nearly 1,100 cores onto a package small enough to fit six times over on a single PCIe card. Now, Esperanto is back, silicon in-hand and taking aim... Read more…

Enter Dojo: Tesla Reveals Design for Modular Supercomputer & D1 Chip

August 20, 2021

Two months ago, Tesla revealed a massive GPU cluster that it said was “roughly the number five supercomputer in the world,” and which was just a precursor to Tesla’s real supercomputing moonshot: the long-rumored, little-detailed Dojo system. “We’ve been scaling our neural network training compute dramatically over the last few years,” said Milan Kovac, Tesla’s director of autopilot engineering. Read more…

CentOS Replacement Rocky Linux Is Now in GA and Under Independent Control

June 21, 2021

The Rocky Enterprise Software Foundation (RESF) is announcing the general availability of Rocky Linux, release 8.4, designed as a drop-in replacement for the soon-to-be discontinued CentOS. The GA release is launching six-and-a-half months after Red Hat deprecated its support for the widely popular, free CentOS server operating system. The Rocky Linux development effort... Read more…

Google Launches TPU v4 AI Chips

May 20, 2021

Google CEO Sundar Pichai spoke for only one minute and 42 seconds about the company’s latest TPU v4 Tensor Processing Units during his keynote at the Google I Read more…

Intel Completes LLVM Adoption; Will End Updates to Classic C/C++ Compilers in Future

August 10, 2021

Intel reported in a blog this week that its adoption of the open source LLVM architecture for Intel’s C/C++ compiler is complete. The transition is part of In Read more…

AMD-Xilinx Deal Gains UK, EU Approvals — China’s Decision Still Pending

July 1, 2021

AMD’s planned acquisition of FPGA maker Xilinx is now in the hands of Chinese regulators after needed antitrust approvals for the $35 billion deal were receiv Read more…

Leading Solution Providers

Contributors

Hot Chips: Here Come the DPUs and IPUs from Arm, Nvidia and Intel

August 25, 2021

The emergence of data processing units (DPU) and infrastructure processing units (IPU) as potentially important pieces in cloud and datacenter architectures was Read more…

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

HPE Wins $2B GreenLake HPC-as-a-Service Deal with NSA

September 1, 2021

In the heated, oft-contentious, government IT space, HPE has won a massive $2 billion contract to provide HPC and AI services to the United States’ National Security Agency (NSA). Following on the heels of the now-canceled $10 billion JEDI contract (reissued as JWCC) and a $10 billion... Read more…

Quantum Roundup: IBM, Rigetti, Phasecraft, Oxford QC, China, and More

July 13, 2021

IBM yesterday announced a proof for a quantum ML algorithm. A week ago, it unveiled a new topology for its quantum processors. Last Friday, the Technical Univer Read more…

Intel Launches 10nm ‘Ice Lake’ Datacenter CPU with Up to 40 Cores

April 6, 2021

The wait is over. Today Intel officially launched its 10nm datacenter CPU, the third-generation Intel Xeon Scalable processor, codenamed Ice Lake. With up to 40 Read more…

Frontier to Meet 20MW Exascale Power Target Set by DARPA in 2008

July 14, 2021

After more than a decade of planning, the United States’ first exascale computer, Frontier, is set to arrive at Oak Ridge National Laboratory (ORNL) later this year. Crossing this “1,000x” horizon required overcoming four major challenges: power demand, reliability, extreme parallelism and data movement. Read more…

Intel Unveils New Node Names; Sapphire Rapids Is Now an ‘Intel 7’ CPU

July 27, 2021

What's a preeminent chip company to do when its process node technology lags the competition by (roughly) one generation, but outmoded naming conventions make it seem like it's two nodes behind? For Intel, the response was to change how it refers to its nodes with the aim of better reflecting its positioning within the leadership semiconductor manufacturing space. Intel revealed its new node nomenclature, and... Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire