Compute Engine Hyperdisk ML throughput reduced
infragcpengineergcp-compute-engine
feature
The minimum provisioned throughput for Hyperdisk ML volumes attached to over 20 instances has been lowered from 100 MiB/s to 20 MiB/s per instance. This change impacts users sharing Hyperdisk ML volumes across many Compute Engine instances, potentially affecting performance tuning. More details are available in the documentation regarding sharing Hyperdisk ML volumes.
Features (1) ›
- Compute Engine Changed
Changed : The minimum provisioned throughput for a Hyperdisk ML volume attached to more than 20 instances is 20 MiB/s per instance, reduced from 100 MiB/s. For more information, see Share a Hyperdisk ML volume between instances .
Read the original announcement →
https://docs.cloud.google.com/release-notes#August_03_2026
Related releases
- Compute Engine flexible CUDs are now GA for G2 and G4 GPU accelerator-optimized machine series Google Cloud release notes ·
- Cloud SDK 580.0.0 Updates Include Deprecation, New Features, and GA Promotions Google Cloud release notes ·
- Critical Conda Channel Bug Fixed in Managed Service for Apache Spark Google Cloud release notes ·
- Malachyte details real-time AI architecture on Google Cloud for retail Google Cloud Blog ·
- Compute Engine machine image disk selection public preview Google Cloud release notes ·
- Cloud SDK 579.0.0: Breaking Changes, Feature Promotions, and New Commands Google Cloud release notes ·