65 ExaFLOP AI Supercomputer being built by AWS and NVIDIA


65 ExaFLOP AI Supercomputer being built by AWS and NVIDIA

Share this article

As the artificial intelligence explosion continues the demand for more advanced artificial intelligence (AI) infrastructure continues to grow. In response to this need, Amazon Web Services (AWS) and NVIDIA have expanded their strategic collaboration to provide enhanced AI infrastructure and services by building a new powerful AI Supercomputer capable of providing 65 ExaFLOPs  of processing power.

This partnership aims to integrate the latest technologies from both companies to drive AI innovation to new heights. One of the key aspects of this collaboration is AWS becoming the first cloud provider to offer NVIDIA GH200 Grace Hopper Superchips. These superchips come equipped with multi-node NVLink technology, a significant step forward in AI computing. The GH200 Grace Hopper Superchips present up to 20 TB of shared memory, a feature that can power terabyte-scale workloads, a capability that was previously unattainable in the cloud.

New AI Supercomputer under construction

In addition to hardware advancements, the partnership extends to cloud services. NVIDIA and AWS are set to host NVIDIA DGX Cloud, NVIDIA’s AI-training-as-a-service platform, on AWS. This service will feature the GH200 NVL32, providing developers with the largest shared memory in a single instance. This collaboration will allow developers to access multi-node supercomputing for training complex AI models swiftly, thereby streamlining the AI development process.

65 ExaFLOP of processing power

The partnership between AWS and NVIDIA also extends to the ambitious Project Ceiba. This project aims to design the world’s fastest GPU-powered AI supercomputer. AWS will host this supercomputer, which will primarily serve NVIDIA’s research and development team. The integration of the Project Ceiba supercomputer with AWS services will provide NVIDIA with a comprehensive set of AWS capabilities for research and development, potentially leading to significant advancements in AI technology. Here are some other articles you may find of interest on the subject of AI supercomputers :

See also  Polestar 4 to be built in South Korea

Summary of collaboration

  • AWS will be the first cloud provider to bring NVIDIA GH200 Grace Hopper Superchips with new multi-node NVLink technology to the cloud. The NVIDIA GH200 NVL32 multi-node platform connects 32 Grace Hopper Superchips with NVIDIA NVLink and NVSwitch technologies into one instance. The platform will be available on Amazon Elastic Compute Cloud (Amazon EC2) instances connected with Amazon’s powerful networking (EFA), supported by advanced virtualization (AWS Nitro System), and hyper-scale clustering (Amazon EC2 UltraClusters), enabling joint customers to scale to thousands of GH200 Superchips.
  • NVIDIA and AWS will collaborate to host NVIDIA DGX Cloud—NVIDIA’s AI-training-as-a-service—on AWS. It will be the first DGX Cloud featuring GH200 NVL32, providing developers the largest shared memory in a single instance. DGX Cloud on AWS will accelerate training of cutting-edge generative AI and large language models that can reach beyond 1 trillion parameters.
  • NVIDIA and AWS are partnering on Project Ceiba to design the world’s fastest GPU-powered AI supercomputer—an at-scale system with GH200 NVL32 and Amazon EFA interconnect hosted by AWS for NVIDIA’s own research and development team. This first-of-its-kind supercomputer—featuring 16,384 NVIDIA GH200 Superchips and capable of processing 65 exaflops of AI—will be used by NVIDIA to propel its next wave of generative AI innovation.
  • AWS will introduce three additional new Amazon EC2 instances: P5e instances, powered by NVIDIA H200 Tensor Core GPUs, for large-scale and cutting-edge generative AI and HPC workloads, and G6 and G6e instances, powered by NVIDIA L4 GPUs and NVIDIA L40S GPUs, respectively, for a wide set of applications such as AI fine-tuning, inference, graphics and video workloads. G6e instances are particularly suitable for developing 3D workflows, digital twins and other applications using NVIDIA Omniverse, a platform for connecting and building generative AI-enabled 3D applications.
  • “AWS and NVIDIA have collaborated for more than 13 years, beginning with the world’s first GPU cloud instance. Today, we offer the widest range of NVIDIA GPU solutions for workloads including graphics, gaming, high performance computing, machine learning, and now, generative AI,” said Adam Selipsky, CEO at AWS. “We continue to innovate with NVIDIA to make AWS the best place to run GPUs, combining next-gen NVIDIA Grace Hopper Superchips with AWS’s EFA powerful networking, EC2 UltraClusters’ hyper-scale clustering, and Nitro’s advanced virtualization capabilities.”
See also  Low-cost thermal camera built using Arduino

Amazon NVIDIA partner

To further bolster its AI offerings, AWS is set to introduce three new Amazon EC2 instances powered by NVIDIA GPUs. These include the P5e instances, powered by NVIDIA H200 Tensor Core GPUs, and the G6 and G6e instances, powered by NVIDIA L4 GPUs and NVIDIA L40S GPUs, respectively. These new instances will enable customers to build, train, and deploy their cutting-edge models on AWS, thereby expanding the possibilities for AI development.

AWS NVIDIA DGX Cloud hosting

Furthermore, AWS will host the NVIDIA DGX Cloud powered by the GH200 NVL32 NVLink infrastructure. This service will provide enterprises with fast access to multi-node supercomputing capabilities, enabling them to train complex AI models efficiently.

To boost generative AI development, NVIDIA has announced software on AWS, including the NVIDIA NeMo Retriever microservice and NVIDIA BioNeMo. These tools will provide developers with the resources they need to explore new frontiers in AI development.

The expanded collaboration between AWS and NVIDIA represents a significant step forward in AI innovation. By integrating their respective technologies, these companies are set to provide advanced infrastructure, software, and services for generative AI innovations. The partnership will not only enhance the capabilities of AI developers but also pave the way for new advancements in AI technology. As the collaboration continues to evolve, the possibilities for AI development could reach unprecedented levels.

Filed Under: Technology News, Top News

Latest aboutworldnews Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, aboutworldnews may earn an affiliate commission. Learn about our Disclosure Policy.

See also  AMD Ryzen 9000 series debuts, built on new Zen 5 architecture

Leave a Reply

Your email address will not be published. Required fields are marked *