Volumes
High-performance NVMe block storage that attaches to VMs and containers.
Why teams use it
Network-attached block storage survives fragile compute.
Volumes are provisioned for VMs, managed databases, and WordPress, use distributed regional block storage, and keep persistent state when workloads restart, fail, or get replaced.
Distributed block service, predictable performance.
Data is striped across a distributed storage plane with automatic healing,designed so a single drive or node loss does not take your database with it.
Latency and throughput scale with the size you purchase, keeping headroom proportional as tables and media libraries grow.
Capacity can grow online where your workload supports it, so midnight resize windows become the exception instead of the rule.
Durable disks
Durable disks survive software deploys and restarts.
Volumes attach to one workload at a time, matching relational databases and stateful services that expect exclusive disk access.
What you get on Day One
Attach durable disks to VMs and containers, grow them when needed, and manage everything from the Tower APIs.
Built for databases
Low-latency distributed block storage suited to MySQL, PostgreSQL, MongoDB, WordPress, and other write-heavy services,not toy file shares.
Grow in place
Increase provisioned size through Tower when your dataset outgrows the original disk,without standing up a parallel volume migration project.
API-first attach
Attach and detach disks to supported workloads from the portal or API so automation and humans stay in sync.
Single-writer semantics
Each volume attaches to one instance at a time,matching how relational databases and many stateful services expect storage to behave.
Configuration & Platform Details
A concise reference for architecture reviews, security questionnaires, and capacity planning.
Platform specifications
- Durability model
- Distributed block storage with automatic healing across the regional storage plane,not a single attached USB disk.
- Access pattern
- Single-writer volumes suited to relational databases and other services that expect exclusive access to a disk.
- Performance
- IOPS and throughput scale with provisioned capacity; pick a size that leaves headroom for growth bursts and maintenance tasks.
- Attachment
- Supported attach targets include Tower Virtual Machines and Container Instances. Mount orchestration is handled by Tower,no direct hypervisor access required.
- Resize
- Online capacity growth where the guest OS and filesystem support extending a live volume.
- Pricing
- Per GB provisioned. Routine attach and detach operations are part of normal use, not a surprise per-call fee.
Pricing
Loading SKUs...