HPC Cloud Updates WE 11 Jan 2026
Updates to AWS, Azure & GCP in the last week relevant for HPC practitioners. A gentle start to the new year by the CSPs but some cool ideas from the nice people at Microsoft and of course some pushing AI into HPC too.
AWS
Agentic CFD? Worth a read and playing with some of the ideas I guess
Accelerating CFD development from years to weeks with agentic AI and AWS | Amazon Web Services
Agentic AI is revolutionizing computational fluid dynamics (CFD) simulations, enabling experienced engineers to focus on physics, innovation, and engineering judgment rather than tedious coding and debugging. Our latest blog explores how this transformative technology can help your team deliver complex projects more rapidly while maintaining scientific rigor.
Spot interruption data now available in Capacity Manager
EC2 Capacity Manager now includes Spot interruption metrics - AWS
Discover more about what’s new at AWS with EC2 Capacity Manager now includes Spot interruption metrics
New regional instances: R8i and flex in Mumbai, Hyberabad and Paris, M8i in Frankfurt and Malaysia, C8i and flex in Mumbai, Seoul and Tokyo
Azure
Azure Cycle Cloud Workspace for Slurm (man Microsoft really sucks at naming) releases a new version
Azure CycleCloud Workspace for Slurm 2025.12.01 Update: Prometheus, Grafana & Entra ID SSO for Scalable HPC
Discover what’s new in Azure CycleCloud Workspace for Slurm 2025.12.01, including Prometheus self‑agent monitoring, managed Grafana dashboards, ARM64 compute…
xpillons
This is interesting and kinda cool, using MPI Broadcast to distribute container images
mpi-stage: High-Performance File Distribution for HPC Clusters | Microsoft Community Hub
When running containerized workloads on HPC clusters, one of the first problems you hit is getting container images onto the nodes quickly and repeatably.…
pauledwards
Google Cloud
I guess Google is still clear out their email inboxes
