Posted On: Aug 17, 2022

We are excited to announce the availability of Amazon EC2 P4d instances in the AWS GovCloud (US) Region. P4d instances are optimized for applications in Machine Learning (ML) and High Performance Computing (HPC).

P4d instances are powered by NVIDIA A100 Tensor Core GPUs and feature second generation Intel® Xeon® Scalable (Cascade Lake) processors, 1.1TB of system memory, and 8TB of local NVMe storage. These instances deliver 400 Gbps instance networking with support for Elastic Fabric Adapter (EFA) and NVIDIA GPUDirect RDMA (remote direct memory access) to enable efficient scale-out of multi-node ML training and HPC workloads.

EC2 P4d instances are optimized for machine learning training workloads such as natural language understanding, perception model training for autonomous vehicles, image classification, object detection and recommendation engines. HPC customers can use P4d’s increased processing performance and GPU memory for applications such as computational fluid dynamics simulations, seismic analysis, drug discovery, DNA sequencing, and financial risk modeling.

Customers can purchase P4d instances as On-Demand Instances, Reserved Instances, Spot Instances, or as part of Savings Plan. To learn more about P4d please visit our product detail page.