Why EC2 Capacity Blocks for ML?
With Amazon Elastic Compute Cloud (Amazon EC2) Capacity Blocks for ML, easily reserve Amazon EC2 P5 instances, powered by the latest NVIDIA H100 Tensor Core GPUs, and Amazon EC2 P4d instances, powered by NVIDIA A100 Tensor Core GPUs, for a future start date. EC2 Capacity Blocks are colocated in Amazon EC2 UltraClusters designed for high-performance machine learning (ML) workloads. You can reserve GPU instances for a duration of one to 14 days and in cluster sizes of one to 64 instances (512 GPUs), giving you the flexibility to run a broad range of ML workloads. EC2 Capacity Blocks can be reserved up to eight weeks in advance.
Benefits
Use cases
-
NVIDIA
-
Arcee
-
Amplify Partners
-
Canva
-
Dashtoon
-
Leonardo.Ai
-
OctoAI
-
Snorkel