Amazon EC2 P6-B200 instances

P6-B200 instances provide 8x NVIDIA Blackwell GPUs with 1440 GB of high bandwidth GPU memory, 5th Generation Intel Xeon Scalable processors (Emerald Rapids), 2 TiB of system memory, and 30 TB of local NVMe storage. These instances with the Blackwell GPUs feature up to 125% increase in GPU TFLOPs, 27% increase in GPU memory size, and 60% increase in GPU memory bandwidth compared to P5en instances.

P6-B200 instances are powered by the AWS Nitro System with specialized hardware and firmware designed to enforce restrictions so that nobody, including anyone in AWS, can access your sensitive AI workloads and data. Nitro live update can deploy firmware updates, bug fixes, and optimizations to Nitro Cards while the system remains operational. This increases stability and reduces downtime, critical to meeting training timelines and running AI applications in production.

To enable efficient distributed training, P6-B200 instances provide 3.2 terabits per second of fourth-generation of Elastic Fabric Adapter networking (EFAv4). These instances are deployed in Amazon EC2 UltraClusters that enable scaling up to tens of thousands of GPUs within a petabit-scale nonblocking network.

P6-B200 instances improve time to train and cost to train, enabling model providers to accelerate time-to-market for bigger and more performant models. P6-B200 instances support a broad range of AI and HPC workloads, from deep learning training and inference to scientific simulations and computer vision applications. They are ideal for medium- to large-scale training and inference workloads.

P6-B200 instances provide up to 8x NVIDIA Blackwell GPUs with 1440 GB of high-bandwidth GPU memory. They boast up to 125% improvement in GPU TFLOPs, 27% increase in GPU memory size and 60% increase in GPU memory bandwidth compared to P5en instances.

P6-B200 instances deliver up to 3.2 terabits per second of EFAv4 networking and 1800 GB/s GPU to GPU interconnect through NVLink. 

P6-B200 instances support Amazon FSx for Lustre file systems so you can access data at the hundreds of GB/s of throughput and millions of IOPS required for large-scale DL and HPC workloads. P6-B200 instances support up to 30 TB of local NVMe SSD storage for fast access to large datasets. You can also use virtually unlimited cost-effective storage with Amazon Simple Storage Service (Amazon S3).

Amazon SageMaker is a fully managed service for building, training, and deploying ML models. With Amazon SageMaker HyperPod (P6-B200 support coming soon), you can more easily scale to tens, hundreds, or thousands of GPUs to train a model quickly at any scale without worrying about setting up and managing resilient training clusters.

AWS Deep Learning AMIs (DLAMI) provides ML practitioners and researchers with the infrastructure and tools to accelerate DL in the cloud, at any scale. AWS Deep Learning Containers are Docker images preinstalled with DL frameworks to streamline the deployment of custom ML environments by letting you skip the complicated process of building and optimizing your environments from scratch.

If you prefer to manage your own containerized workloads through container orchestration services, you can deploy P6-B200 instances with Amazon Elastic Kubernetes Service (Amazon EKS) or Amazon Elastic Container Service (Amazon ECS).