Zoned block device (ZBD)
A zoned block device exposes a logical block address space divided into zones with zone-specific access rules.
The device model is not Linux itself. Linux's zoned block layer provides a common interface to supported devices. Conventional, sequential-write-required and sequential-write-preferred zones are different types; the permitted types depend on the device model and protocol. ZNS and host-managed SMR are related implementations, not synonyms.
Do not infer this access model from zoned bit recording. ZBR describes the HDD's radial recording layout, not sequential-write rules. Historical SCSI Notch pages could expose recording geometry without defining this modern zone contract. [6] [2]
Each sequential zone has a write pointer marking the next write position, and its data cannot be overwritten in place until a zone reset returns the pointer to the zone start. Conventional zones have no write pointer and accept random writes. Sequential-write-required zones occur only on host-managed devices, and sequential-write-preferred zones, which keep a pointer yet accept random writes, only on host-aware devices. [2]
Zone capacity, the usable blocks from the zone start, is at most the zone size. In a documented example, a 15 TB host-managed shingled disk with 256 MB zones exposes each sequential zone as a file whose maximum size, 524288 blocks of 512 B, equals the zone capacity. [3]
Relations
- related
- Host-managed SMR (HM-SMR), Zoned namespaces (ZNS), Shingled magnetic recording (SMR), Zone capacity, Zone Append, Zone write plugging, Zone write locking, Host-side flash GC, ZAC, Zone Reset, Open-zone limit, Zoned bit recording (ZBR), Zone Finish, Block device
Advantages
- Separating zone capacity from zone size keeps a constant zone size while letting capacity follow the medium, such as flash erase blocks. [2]
Limitations
- Rotational zoned devices default to writes from a single issuer at queue depth 1, which may not suit remote devices on high-latency networks. [7]
Citations
- Western Digital, Zoned Storage: device models, zone types, management commands and Zone Append.
- Linux kernel documentation, zonefs: sequential files, explicit-open and runtime sysfs attributes.
- Seagate: ST3655 Family SCSI Product Manual, Rev. A, 18 January 1994, sections C.3 and C.7, pp. 97-98 and 108-109
- Linux kernel documentation: stable ABI, sysfs-block queue attributes zoned, nr_zones and zoned_qd1_writes.
Further reading
- Zoned block device (ZBD)
- Hesela: ZNS resource limits and worked examples
- Western Digital, Zoned Storage: Write Ordering Control; locking through Linux 6.9 and plugging from Linux 6.10.
Sources & caveats
- zoned-overview re-read 2026-10-23 for write pointers, zone reset, zone capacity and the behaviour of the three zone types.
- zonefs re-read 2026-10-23 for the 15 TB, 256 MB zone example and the 524288 x 512 B file size.
- linux-sysfs-block read 2026-10-23 for the zoned_qd1_writes queue attribute behind the queue-depth limitation.
- The zone-finish entry links here because Zone Finish is a state transition defined by the zone model this entry describes.
- A companion analysis is listed but not used in prose.