Block Storage

Block storage divides data into fixed-size chunks called blocks and presents them to a server as a raw volume, much like a physical hard drive. The operating system formats the volume with a file system or uses it directly for a database. In the cloud, block storage volumes attach to virtual machines and deliver low latency and high throughput.

How It Works

A block volume appears to the host as an unformatted disk. The host manages how blocks map to files. Cloud block storage, such as Amazon EBS, Azure Managed Disks, or Google Persistent Disk, lets customers choose volume types tuned for capacity, throughput, or input/output operations per second (IOPS). Volumes can be resized, snapshotted for backup, encrypted, and detached from one instance and reattached to another. Most block volumes attach to a single instance at a time, although some support multi-attach. Performance and durability come from replicating blocks within the provider's infrastructure.

Why It Matters

Block storage is the right choice for workloads that need fast, consistent, low-latency access and the ability to modify data in place: relational and NoSQL databases, transactional applications, boot volumes, and file servers. It complements object storage, which is cheaper and more scalable but higher latency and write-whole. Choosing the correct volume type and IOPS level is a key performance and cost decision, since over-provisioning wastes money and under-provisioning throttles the workload.

Related Terms

Block storage contrasts with object storage, underpins databases, and is provisioned through infrastructure-as-a-service.