File System Support for Zoned Block Devices
Zoned block device (ZBD) support has been introduced in Linux with kernelversion 4.10. ZBDs have different write constraints than regular block devices.A ZBD is divided into several zones and each zon …
| Talk Title | File System Support for Zoned Block Devices |
| Speakers | Naohiro Aota (Staff Engineer, Western Digital), Damien Le Moal (Western Digital) |
| Conference | Open Source Summit + ELC Europe |
| Conf Tag | |
| Location | Lyon, France |
| Date | Oct 27-Nov 1, 2019 |
| URL | Talk Page |
| Slides | Talk Slides |
| Video | |
Zoned block device (ZBD) support has been introduced in Linux with kernelversion 4.10. ZBDs have different write constraints than regular block devices.A ZBD is divided into several zones and each zone must be written sequentially.The main type of ZBD currently available is SMR HDDs. The NVMe Zoned NameSpaceproposal is also being drafted to add a zone abstraction to the NVMe specifications.Natively supporting ZBDs in a filesystem is not a trivial change. Somefilesystems must rely on special block layer drivers to ensure sequential writes(e.g. ext4 and the dm-zoned device mappers). Filesystems using a copy-on-writedesign are better candidates for native ZBD support. Examples are F2FS and btrfs.This talk discusses the principles of ZBD native support in filesystems. Supportin F2FS is discussed and the approach taken with btrfs is next presented. Thisis followed with a performance comparison between filesystems with native ZBDsupport and regular ones using dm-zoned.
