Channel Insider content and product recommendations are editorially independent. We may make money when you click on links to our partners. View our editorial policy here.
The infrastructure arms race for AI continues, and AWS just made a major move. The cloud giant has announced the general availability of its P6e-GB200 UltraServers, powered by Nvidia Grace Blackwell Superchips. These new systems are engineered to handle the most demanding AI workloads on the planet—from trillion-parameter model training to real-time reasoning systems and next-gen agentic AI.
“The scale of the AI systems that our customers are building today—across drug discovery, enterprise search, software development, and more—is truly remarkable,” said David Brown, VP of compute and machine learning at AWS. “With the new Blackwell systems, we’re providing secure, reliable GPU infrastructure at a massive scale, so that customers can confidently push the boundaries of AI.”
Extreme Performance Meets Efficient Cooling
Each UltraServer is a powerhouse of up to 72 Blackwell GPUs linked by Nvidia’s latest NVLink interconnect, acting as a single compute unit. This impressive configuration results in a staggering 360 petaflops of FP8 compute, 13.4 TB of high-bandwidth memory, and 28.8 Tbps of Elastic Fabric Adapter (EFAv4) networking. Such specifications empower users to engage in training and inference operations that were previously considered unattainable.
What distinguishes these UltraServers isn’t merely their brawn; they also incorporate AWS’s first large-scale liquid cooling system with a custom-built In-Row Heat Exchanger (IRHX). “Existing commercial cooling solutions either consumed too much floor space or couldn’t scale to our needs,” Brown explains. So, in true innovative spirit, AWS created a solution tailored specifically for their ambitious goals.
Two Blackwell Options, Built for Different Scales
The P6e-GB200 joins the previously launched P6-B200 instance family, which features 8 Blackwell GPUs and offers flexibility for medium-to-large scale AI and high-performance computing workloads. Together, these instances form the bedrock of AWS’s evolving AI infrastructure strategy.
Brown highlights that the “P6-B200 instances are a great fit for customers looking to migrate existing workloads with minimal code changes.” Meanwhile, the P6e-GB200 is aimed at those pushing the frontiers of AI, enabling performance and efficiency levels that were once out of reach for many enterprises.
Tightly Integrated with AWS Services and Security
But it’s not just about raw horsepower; AWS has embedded these innovative systems within its larger managed ecosystem. This includes Amazon SageMaker HyperPod, which streamlines the provisioning and management of extensive GPU clusters, and Amazon EKS, now optimized for the GB200 NVL72 topology and GPU-aware scheduling. Furthermore, the P6e-GB200 UltraServers will be accessible through NVIDIA DGX Cloud, granting users entry to Nvidia’s full AI software stack and expert support.
AWS emphasizes that security, reliability, and scalability are pivotal differentiators. Both the P6e and P6-B200 systems operate on the Nitro System, AWS’s secure compute foundation that enables live patching, fine-grained isolation, and continuous optimization without any downtime.
In a landscape where AI workloads are advancing exponentially, AWS is clearly wagering that robust infrastructure will be crucial for partners to maintain their competitive edge. By focusing on building a framework that accommodates increasingly sophisticated AI workloads, AWS positions itself for long-term success. The Blackwell-powered UltraServers give Managed Service Providers (MSPs) a sturdy foundation, enabling them to expand their AI services portfolios and effectively support their clients throughout the entire AI adoption journey.
“This launch announcement is an important milestone, and it’s just the beginning,” affirm Brown. “As AI capabilities evolve rapidly, you need infrastructure built not just for today’s demands, but for all the possibilities that lie ahead. With innovations across compute, networking, operations, and managed services, P6e-GB200 UltraServers and P6-B200 instances are ready to enable these possibilities. We can’t wait to see what you will build with them.”
Recently, AWS announced new initiatives to change how partners engage with AWS and provide enhancements that focus on helping partners grow their relevance, reach, and relationships with their customers.