June 28, 2021 – Supercomputing centers around the world looking for a flexible and energy-efficient alternative to diversify their systems are starting to look to Arm for their exascale supercomputers.

Arm is the most popular CPU architecture in the world, according to the company, but the x86 architecture has over 97% of the data center market, thanks to its ecosystem of partners, software and development tools.

NVIDIA has a long tradition of supporting all CPU architectures, including x86 and POWER from IBM. NVIDIA also supports the Arm architecture. It is a job that we want to significantly develop and accelerate by acquiring the company to create a unique combination for our partners and customers.

“Arm’s flexible and energy-efficient architecture is ideal for all types of workloads, including high performance computing, cloud and edge applications. NVIDIA’s efforts to accelerate Arm applications will fuel innovation, strengthen the ecosystem, and provide greater choice for customers and end users, ”said Rene Haas, president of Arm’s intellectual property group, during a recent conference at GTC.

Supercomputer arms for Exascale

Arm’s open architecture gives businesses the flexibility to design processors optimized for their workloads. High performance data centers around the world are taking advantage of this powerful option.

“As a supplier of the high performance Arm-based processor designed under the European Processor Initiative to power European exascale supercomputing, SiPearl is committed to advancing the Arm ecosystem,” said Philippe Notton, Founder and CEO by SiPearl.

“Implementing our CPU in GPU-accelerated nodes requires a solid development environment, domain-specific libraries, development kits, and HPC application tuning. NVIDIA’s investment and expertise in all of these areas is driving sustained ecosystem and market growth, ”he added.

ETRI, a national research institute in South Korea, is developing an Arm-based processor for its future supercomputer. It will support double precision and mixed precision math to tackle HPC and AI applications.

And the Indian C-DAC is building an Arm-based processor to power an exascale system for its scientific research and AI initiatives.

NVIDIA Hands Arm New Tools

These are just a few of the many Arm-based HPC initiatives that can leverage the NVIDIA HPC SDK, a comprehensive suite of compilers, libraries, and tools that simplify application development and porting to architecture. Arm. The SDK serves as the foundation for an accelerated Arm HPC ecosystem.

The NVIDIA HPC SDK is used to develop and port HPC applications to Arm.

The next software release – HPC SDK 21.7, coming in July – will provide more Arm intrinsics, features that compilers can use to tune performance. And it provides custom math functions specially optimized for Arm processors.

Additionally, NVIDIA plans to support scalable vector extensions on Arm’s Neoverse platform. SVE debuted in the Fujitsu A64FX that powers Fugaku, ranked # 1 on the TOP500 list of the world’s fastest supercomputers.

Accelerated arm kit coming in July

We also facilitate the creation, evaluation and benchmarking of HPC and AI applications on Accelerated Arm systems with the NVIDIA Arm HPC Development Kit. This is a platform available from NVIDIA and GIGABYTE as software loaded on a server powered by a processor based on Ampere Altra Arm, NVIDIA A100 Tensor Core GPUs and NVIDIA BlueField-2 DPUs. for accelerated networking.

Developers from more than 70 leading organizations, including the Los Alamos National Laboratory, have requested early access to the kit, which will ship in July.

“We are building some of our key applications as well as some AI / ML workflows for the Arm architecture using the NVIDIA HPC SDK,” said Steve Poole, Chief Next Generation Systems Architect at Los Alamos. “NVIDIA Arm Development Kits will make it easier to transition our codes to NVIDIA GPUs and Arm processors,” he said.

Arm researchers with apps

NVIDIA also offers versions for the Arm architecture of popular HPC applications in molecular dynamics (NAMD, Tinker-HP), materials modeling (Quantum Espresso) and other areas.

Applications are available as containers in NVIDIA NGC, a software registry optimized for the GPU. These containers simplify the deployment of applications on Arm-based systems, so researchers can focus on advancing their science.

Extend the arm with more partners

The power of Arm processors combined with NVIDIA GPUs meet high performance computing needs beyond supercomputing centers.

For example, NVIDIA GPUs and Ampere Altra CPUs from Ampere Computing serve the data center to cloud markets. In edge computing, NVIDIA is working with Marvell Semiconductor to combine its OCTEON Arm processors with NVIDIA GPUs to accelerate AI workloads for network optimization and security.

Our collaborations also include system suppliers. GIGABYTE led the charge by delivering Accelerated Arm Servers.

“We are excited to be part of the Arm ecosystem with a portfolio of Arm servers powered by NVIDIA GPUs and DPUs and Ampere Altra CPUs,” said Etay Lee, CEO of GIGABYTE.

“We are committed to working with NVIDIA to grow the Arm ecosystem, from hardware and software development tools to Arm applications, helping our customers accelerate their HPC workloads across all industries,” he said. added.

More on the Horizon

We’re also expanding the NVIDIA Certified Program of nearly 40 systems from a dozen OEMs to include Arm-powered systems, so organizations have more choices in the types of pre-tested systems they can deploy with confidence.

GIGABYTE and Wiwynn plan to offer servers with Arm Neoverse processors with NVIDIA Ampere GPUs, BlueField-2 DPUs, or both. The servers are expected to be available next year and will be subject to NVIDIA certification as they are released.

We’re only scratching the surface of what’s possible with accelerated, energy-efficient computing powered by the combination of Arm, NVIDIA and our partners.


Source: PARESH KHARYA, NVIDIA



Source link

About The Author

Related Posts