AWS Parallel Computing Service
Managed Slurm HPC
A managed HPC service (AWS PCS): AWS operates the Slurm controller, while you define compute node groups on EC2 and submit jobs just like on a familiar HPC cluster.
- Cost$$ Medium
- Difficulty●●●●●
- Costs money when idle?Yes
- Managed by AWS
Think of it asLike renting a computer lab with an admin on duty: you just queue up and submit your work.
When to use it
✓ Use it when
You want to run HPC with Slurm without managing or upgrading the scheduler yourself.
✗ Avoid it when
You need deep customization of the cluster or scheduler → ParallelCluster; simple container jobs → Batch.
Common uses
- Engineering simulation (CFD, FEA)
- Scientific research using Slurm
- Moving on-prem HPC clusters to the cloud
What does it cost?
- Pricing model: Hourly fee for the Slurm controller (by cluster size) and a per-node management fee, plus the EC2 and storage used.
