Amazon Web Services (AWS) and NVIDIA have unveiled the integration of NVIDIA's latest GPU platform, the NVIDIA Blackwell GPU, with AWS offerings, ushering in a new era of advanced infrastructure for generative artificial intelligence (AI) applications.
This collaboration leverages NVIDIA's GPU technology within AWS's robust infrastructure and services ecosystem, including the NVIDIA GB200 Grace Blackwell Superchip and B100 Tensor Core GPUs. By combining their strengths, the two companies aim to provide customers with secure and highly advanced solutions for unlocking the full potential of generative AI.
The introduction of the NVIDIA Blackwell platform on AWS marks a significant milestone in their longstanding partnership. It brings together a range of technologies, such as AWS's Nitro System, Elastic Fabric Adapter (EFA) networking, and Amazon EC2 UltraClusters, with NVIDIA's powerful GPU capabilities. This integration enables customers to accelerate real-time inference on multi-trillion parameter large language models (LLMs) more efficiently and cost-effectively than ever before.
“The deep collaboration between our two organizations goes back more than 13 years, when together we launched the world’s first GPU cloud instance on AWS, and today we offer the widest range of NVIDIA GPU solutions for customers. NVIDIA’s next-generation Grace Blackwell processor marks a significant step forward in generative AI and GPU computing. When combined with AWS’s powerful Elastic Fabric Adapter Networking, Amazon EC2 UltraClusters’ hyper-scale clustering, and our unique Nitro system’s advanced virtualization and security capabilities, we make it possible for customers to build and run multi-trillion parameter large language models faster, at massive scale, and more securely than anywhere else. Together, we continue to innovate to make AWS the best place to run NVIDIA GPUs in the cloud,” said Adam Selipsky, CEO at AWS.
NVIDIA Blackwell platform
One of the key offerings is the NVIDIA Blackwell platform, which includes the GB200 NVL72—a superchip equipped with 72 Blackwell GPUs and 36 Grace CPUs interconnected by fifth-generation NVIDIA NVLink. When combined with AWS's networking and clustering capabilities, customers can scale up to thousands of GB200 Superchips, significantly enhancing the speed and efficiency of inference workloads for resource-intensive language models.
Additionally, AWS plans to deploy EC2 instances featuring the new B100 GPUs in EC2 UltraClusters, further optimizing generative AI training and inference at a massive scale. These advancements build upon the success of previous NVIDIA-powered EC2 instances, demonstrating a commitment to continuous innovation and improvement.
Security is the main focus
Security remains a top priority in AI deployments, and AWS and NVIDIA have taken significant steps to protect customer data and model integrity. The combination of AWS's Nitro System and NVIDIA's GB200 offers enhanced security features, including physical encryption of connections and data transfer, as well as encryption capabilities for distributed training and inference. In addition, AWS Nitro Enclaves and AWS Key Management Service (KMS) provide customers with unprecedented control over their data, ensuring that it remains secure throughout the training process.
Going beyond infrastructure updates
The collaboration between AWS and NVIDIA extends beyond infrastructure enhancements to include joint efforts in AI research and development. Project Ceiba, a supercomputer hosted exclusively on AWS, will leverage the new NVIDIA Blackwell platform to accelerate research in AI innovation. Additionally, AWS and NVIDIA are expanding their focus on healthcare and life sciences, offering advanced generative AI applications and microservices to drive breakthroughs in drug discovery and medical research.
"AI is driving breakthroughs at an unprecedented pace, leading to new applications, business models, and innovation across industries. Our collaboration with AWS is accelerating new generative AI capabilities and providing customers with unprecedented computing power to push the boundaries of what's possible," said Jensen Huang, Founder and CEO of NVIDIA.
In other news, Amazon Web Services (AWS) has unveiled plans to launch a new infrastructure region in Mexico, reinforcing its commitment to meeting the rising demand for cloud services in Latin America.
