Zonefs: Features Roadmap
summary
Zonefs exposes each ZNS zone as an append‑only file and already fixes a read‑ahead infinite loop and enforces device limits via sysfs. Planned changes include changing inode locking so finish doesn’t stall concurrent reads, replacing per‑zone inodes with a report_zones cache to drop memory from ~128 bytes to about 16 bytes per zone, enabling async zone append (io_uring large CQ support) and adding buffered writes via controlled pagecache copies to remove the strict O_DIRECT-only path.
AI-written summary · about
chapters (7)
- 0:00Introduction to Zonefs
- 1:50Zonefs basics and limitations
- 3:58Recent updates and fixes
- 7:50Future improvements planned
- 16:01Asynchronous I/O and append support
- 18:52Buffered writes and O_DIRECT
- 21:43Q&A and closing
related talks
31mFrom Open-Channel SSDs to Zoned Namespaces
14mZNS+: Advanced Zoned Namespace Interface for Supporting In-Storage Zone Compaction
14mUnleashing Zoned UFS: Cross-Layer Optimizations for Next-Generation Mobile Storage
25mUsing Zoned Block Devices in Fedora
44mReinventing the File System: OpenZFS
44mThe ZFS filesystem
20mZNSwap: un-Block your Swap
38mRethinking Writeback: Scaling Linux Filesystem and Memory Performance for the Next Decade
