Western Digital Has Been Developing A New Linux File-System: Zonefs

Western Digital has been contributing a lot more to the Linux kernel in recent years from RISC-V architecture bits to storage enhancements. The most recent code they have been working on in recent weeks is a brand new Linux file-system.

But before getting too bent out of shape over yet-another-Linux-filesystem, the new Western Digital creation isn't intended to be a general purpose file-system for competing with the likes of EXT4, Btrfs, XFS, and ZFS On Linux... This new file-system, Zonefs, is for specialty use-cases and running on zoned block devices. Zonefs exposes each zone of a zoned block device as a file, compared to traditional file-systems or how zoned block device support is exposed through the likes of F2FS and friends on host-managed/host-aware SMR (Shingled Magnetic Recording) disk drives.

Damien Le Moal of Western Digital describes Zonefs as "zonefs is not a POSIX compliant file system. It's goal is to simplify the implementation of zoned block devices support in applications by replacing raw block device file accesses with a richer file based API, avoiding relying on direct block device file ioctls which may be more obscure to developers. One example of this approach is the implementation of LSM (log-structured merge) tree structures (such as used in RocksDB and LevelDB) on zoned block devices by allowing SSTables to be stored in a zone file similarly to a regular file system rather than as a range of sectors of a zoned device. The introduction of the higher level construct "one file is one zone" can help reducing the amount of changes needed in the application while at the same time allowing the use of zoned block devices with various programming languages other than C."

Western Digital has been investing big into zoned storage for allowing larger capacity economical devices. More background information on zoned storage can be found via the Western Digital blog.

Their proposed Zonefs file-system is currently coming in at only about fifteen-hundred lines of new code considering it's not a complete file-system but just a specialty implementation for better exposing zoned storage.