Generally Available: Azure Ephemeral OS Disk with full caching for VM/VMSS

Azure Ephemeral OS Disk with full caching is now generally available for new virtual machines and Virtual Machine Scale Sets. The capability caches the complete OS image on a VM’s local storage, eliminating remote-storage reads after caching finishes. This delivers consistently fast, low-millisecond OS disk access, up to 10× better I/O performance, and greater resilience during storage interruptions—without slowing VM provisioning because caching runs asynchronously after boot. It is designed for I/O-sensitive, stateless workloads such as AI training and inference, quorum-based databases, real-time analytics, and large-scale services. Full caching uses a VM’s cache, resource, or NVMe disk and reduces available local capacity by twice the OS disk size.

It is available across Azure public regions for supported 8-core-and-larger GPU, storage-optimized, memory-optimized, HPC, general-purpose, and compute-optimized VM series. Customers can enable it through Azure CLI, ARM templates, or REST APIs by setting the full-caching option for Ephemeral OS disks.