Block Storage

Block Storage is a storage architecture in which data is split into blocks of fixed size and stored in an individually addressable way. Operating systems and applications thereby obtain very direct, granular access to the storage.

Block storage is typically provided as volumes or LUNs via a storage area network (SAN) with protocols such as Fibre Channel, iSCSI or, increasingly, NVMe over Fabrics. The file system resides on the side of the accessing system – the storage itself knows no directories. This is precisely what makes block storage the first choice for databases, virtual machines and other workloads with high demands on latency and consistent performance.

In contrast to file storage (hierarchical directories, shared access) and object storage (flat structure, API access, maximum scaling), block storage is designed for system-level, performance-critical applications. Many business-critical systems in the enterprise environment are based on precisely this form of storage – the choice of architecture always follows the use case.