AWS HPC
EC2 HPC instances (hpc7a, hpc7g, hpc6a), P5 H100 and P5en H200 GPU instances on EFA, the free ParallelCluster tool and AWS Parallel Computing Service, a managed Slurm controller.
Facts
Source per value- Type What kind of provider it is: hyperscaler, managed HPC service, GPU cloud, sovereign or EU-hosted cloud, publicly funded supercomputer, or simulation-software cloud.
- hyperscaler 1
- Fabric Primary node-to-node network the provider states for its HPC or GPU clusters. RDMA means the provider names RDMA cluster networking without naming the transport.
- EFA 1
- Interconnect detail Speeds and instance families as the provider states them.
- EFA: 300 Gbps on hpc7a, 200 Gbps on hpc7g, 100 Gbps on hpc6a, up to 3,200 Gbps on P5/P5e/P5en 1
- Schedulers Job schedulers and orchestrators the provider offers or documents.
- AWS Parallel Computing Service (managed Slurm), AWS ParallelCluster (open source), AWS Batch 5
- Slurm Offers or documents Slurm (managed, or as a supported deployment).
- Yes 5
- Kubernetes Offers managed Kubernetes for its clusters.
- Not published
- CPUs CPU models the provider names for its HPC or GPU hosts.
- hpc7a: 4th Gen AMD EPYC, up to 192 cores; hpc7g: AWS Graviton3E, 64 cores; hpc6a: 3rd Gen AMD EPYC, 96 cores 1
- GPUs / accelerators Accelerators the provider lists, as listed on its own pages.
- NVIDIA H100 (P5), NVIDIA H200 (P5en), 8 per instance 4
- H100 Lists NVIDIA H100.
- Yes 4
- Blackwell Lists at least one NVIDIA Blackwell data-centre GPU (B200, B300, GB200, GB300).
- Not published
- AMD Instinct Lists at least one AMD Instinct GPU.
- Not published
- Bare metal Sells non-virtualised bare-metal nodes.
- Not published
- H100 $/GPU-hr $/GPU-hr Published per-GPU-hour price of an H100. Basis (on-demand, capacity block, derived from an 8-GPU price) is in the next column.
- $6.88 8
- H100 price basis What the H100 price is: purchase mode, instance size and any derivation.
- On-demand p5.4xlarge (1x H100) Linux in us-east-1 at $6.88 per instance-hour; the 8-GPU p5.48xlarge is $55.04 ($6.88 per GPU). EC2 Capacity Blocks run $5.97 per GPU-hour ($47.757 per p5.48xlarge hour). 8
- Price checked Date the price was read from the provider page.
- 2026-11-11 8
- Pricing model How the provider charges, as it states it.
- On-demand and Capacity Blocks for EC2; PCS adds a controller fee and a per-node fee ($0.08 per instance-hour on standard types); ParallelCluster is free 7
- Regions Locations the provider names on its own pages.
- Not published
- EU-hosted Names data-centre capacity in an EU member state.
- Not published
- Sovereignty claim The provider's own sovereignty or compliance statement, quoted or closely paraphrased.
- Not published
- Software catalogue Pre-configured application catalogue a platform provider advertises.
- Not published
- Runs on Clouds and hardware partners a platform provider lists.
- Not published
- Pricing model
- On-demand, Capacity Blocks
- From
- $6.88 GPU-hour (H100, on-demand)
- Note
- p5.4xlarge (1x H100) on-demand $6.88/hr and p5.48xlarge (8x H100) $55.04/hr in us-east-1. Capacity Blocks: $47.757 per p5.48xlarge hour ($5.97 per GPU-hour). p5en.48xlarge (8x H200) on-demand $63.296/hr.
Sources
- EC2 Hpc7a instances accessed 2026-11-11
- EC2 Hpc7g instances accessed 2026-11-11
- EC2 Hpc6a instances accessed 2026-11-11
- EC2 P5 instances accessed 2026-11-11
- AWS Parallel Computing Service accessed 2026-11-11
- AWS ParallelCluster accessed 2026-11-11
- AWS PCS pricing accessed 2026-11-11
- Amazon EC2 On-Demand pricing (US East, N. Virginia, Linux) accessed 2026-11-11
- EC2 Capacity Blocks pricing accessed 2026-11-11
Related providers and lists
Get a quote from AWS HPC and similar providers
Tell us what you need once. We send it to matching providers and they reply to you directly.