> **For AI agents:** The documentation index is at [https://docs.digitalocean.com/llms.txt](https://docs.digitalocean.com/llms.txt). Markdown versions of pages use the same URL with `index.html.md` in place of the HTML page (for example, append `index.html.md` to the directory path instead of opening the HTML document).

# Volume Features

Volumes are network-attached block storage. You can use them with Droplets or Kubernetes clusters, move or resize them, and create snapshots at any time.

DigitalOcean Volumes provide flexible, high-performance block storage that you can attach to [Droplets](https://docs.digitalocean.com/products/droplets/index.html.md) or use with [DigitalOcean Kubernetes clusters](https://docs.digitalocean.com/products/kubernetes/index.html.md) to persist data without adding extra CPU or memory. Volumes support a wide range of workloads, including web servers, databases, backups, and AI/ML workflows.

When attached to a Droplet, Volumes function like local block devices. You can partition, format, and mount them using standard disk tools. DigitalOcean also supports automatic formatting and mounting for first-time use on Ubuntu, Fedora, Debian 8+, CentOS, and Fedora Atomic.

In Kubernetes clusters, you can attach Volumes to pods using persistent volumes to store data beyond the pod lifecycle. For details, see [Add Volumes to DigitalOcean Kubernetes](https://docs.digitalocean.com/products/kubernetes/how-to/add-volumes/index.html.md).

## Key Capabilities

Volumes are independent resources, which gives you flexibility to:

- [Move volumes between Droplets](https://docs.digitalocean.com/products/volumes/how-to/mount-unmount/index.html.md) in the same datacenter without downtime.
- [Increase a volume’s size](https://docs.digitalocean.com/products/volumes/how-to/increase-size/index.html.md) at any time without powering off the attached Droplet.

Volumes also support [snapshots](https://docs.digitalocean.com/products/snapshots/index.html.md), which you can use to:

- [Create volume snapshots](https://docs.digitalocean.com/products/snapshots/how-to/snapshot-volumes/index.html.md) on demand to save a full copy of the volume’s contents.
- [Create new volumes from snapshots](https://docs.digitalocean.com/products/snapshots/how-to/create-volumes/index.html.md) to clone an environment or replicate data.

## Use Cases

Volumes are most useful when you need additional storage capacity without resizing your Droplet. Common use cases include:

- The document root or media directory for a web server.
- Database data directories.
- A target for backups or archived data.
- Expanded storage for personal file hosting tools like ownCloud.
- Storage for distributed web applications or AI/ML pipelines that need scalable, persistent block storage.

## Performance

Volumes use SSD-backed storage to provide consistent IOPS and throughput for most workloads.

They also include burst support, which temporarily increases IOPS and bandwidth for up to 60 seconds when workloads spike. Burst support automatically cools down over the next 60 seconds before resetting, helping short, intense operations complete faster.

| Type | IOPS | Throughput |
|---|---|---|
| Shared | 7,500 | 300 MB/s |
| Shared (burst) | 10,000 | 450 MB/s |
| Dedicated | 10,000 | 450 MB/s |
| Dedicated (burst) | 15,000 | 525 MB/s |
| | | |

## Security and Durability

All DigitalOcean Volumes are encrypted with LUKS (Linux Unified Key Setup). The storage cluster is fully encrypted at rest, and any snapshots you create inherit this encryption.

Volumes are built on Ceph, a distributed storage system designed for durability and high availability. Ceph replicates data across multiple hosts, automatically heals from hardware failures, and uses continuous monitoring to maintain performance and reliability. For more detail on our design decisions, see our engineering blog post, [Why We Chose Ceph to Build Block Storage](https://www.digitalocean.com/blog/why-we-chose-ceph-to-build-block-storage).

Your data is stored on hardware outside any individual Droplet and replicated across multiple hosts. This reduces the risk of data loss from hardware failures and provides a durable, highly available storage layer for your workloads.