HPC Cloud Updates WE 12 Apr 2026

Updates to AWS, Azure & GCP in the last week relevant for HPC practitioners. A new Azure region, a filesystem over AWS S3 and running GPUs at scale from Google.

HPC Cloud Updates WE 12 Apr 2026

AWS

Perhaps not strictly HPC but I’m already aware of people using S3 (and other BLOB storage exposing an S3 API) in HPC workloads that I’m fairly sure that won’t be long before we see HPC running on top of this too

Launching S3 Files, making S3 buckets accessible as file systems | Amazon Web Services
Amazon S3 Files makes S3 buckets accessible as high-performance file systems on AWS compute resources, eliminating the tradeoff between object storage benefits and interactive file capabilities while enabling seamless data sharing with ~1ms latencies.

Maybe a little more HPC. NetApp is available in London, Hyderabad and Sao Paolo now

Second-generation Amazon FSx for NetApp ONTAP is now available in four additional AWS commercial and AWS GovCloud (US) Regions - AWS
Discover more about what’s new at AWS with Second-generation Amazon FSx for NetApp ONTAP is now available in four additional AWS commercial and AWS GovCloud (US) Regions

This might help allocate costs to your HPC clients if you’re running a platform with multiple users

Amazon EC2 Capacity Manager now supports tag-based dimensions - AWS
Discover more about what’s new at AWS with Amazon EC2 Capacity Manager now supports tag-based dimensions

New Regional Instances:  X8i in Paris


Azure

New Azure region in Denmark

Azure updates | Microsoft Azure
Subscribe to Microsoft Azure today for service updates, all in one place. Check out the new Cloud Platform roadmap to see our latest product plans.

I guess if you’re trying to run LLM inference using K8 this could be relevant. 

Azure updates | Microsoft Azure
Subscribe to Microsoft Azure today for service updates, all in one place. Check out the new Cloud Platform roadmap to see our latest product plans.

Google Cloud

Google gave us a guide on building GPU infrastructure

A guide to architecting reliable GPU infrastructure | Google Cloud Blog
Understand the core challenges behind reliable GPU-based infrastructure on Google Cloud, and find resources to dive deeper.

but also claims of better carbon efficient of its Ironwood TPUs

Ironwood TPUs deliver 3.7x carbon efficiency gains | Google Cloud Blog
The seventh-generation Ironwood TPU achieved an approximately 3.7x improvement in Compute Carbon Intensity (CCI) compared to TPU v5p architecture.