AWS Parallel Computing Service supports P6e-GB200 and P6e-GB300 UltraServers
News
AWS Parallel Computing Service (PCS) now supports Amazon EC2 P6e-GB200 and P6e-GB300 UltraServer instances with NVIDIA Blackwell architecture for large-scale GPU workloads.
- P6e-GB200 provides up to 72 NVIDIA Blackwell GPUs with 360 petaflops FP8 compute and 13.4 TB HBM3e memory
- P6e-GB300 offers 1.5x GPU memory and 1.5x FP4 compute compared to P6e-GB200
- Reserve UltraServers through EC2 Capacity Blocks for ML and associate via launch templates
- PCS automatically configures Slurm with correct topology plugin for cluster management
- Available in all AWS Regions where PCS and EC2 Capacity Blocks for UltraServers are supported
AWS PCS simplifies running and scaling HPC workloads with managed cluster operations, built-in observability, and integrated compute, storage, and networking tools.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
Aug 5
2025
2025
AWS Parallel Computing Service now supports Internet Protocol Version 6 (IPv6)
Aug 28
2024
2024
Announcing AWS Parallel Computing Service
Mar 27
2026
2026
AWS Parallel Computing Service supports slurmdbd and cgroups settings
Aug 28
2024
2024
Announcing AWS Parallel Computing Service to run HPC workloads at virtually any scale
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.