The second Release Candidate of the upcoming version.
Tag: OpenZFS
-
OpenZFS continues to evolve as a robust filesystem for everything from IoT devices to supercomputing clusters. The upcoming OpenZFS 2.4 release (Nov 2025) focuses on stability, usability, and performance.
Key Features in 2.4
- Fast dedup log pacing
- Fix for Encryption + ZFS send
- Skip slow RAID-Z children
- Improved allocation under fragmentation
- Scrub only changed blocks in a date range
- Parallel ARC eviction for large-memory systems
- Asynchronous ARC flush on pool export
- New JSON output,
zfs rewritecommand, project quotas on FreeBSD, ZVOL threading, and more
In-Development Features
- Label Redesign: Larger (256 MiB) labels, more uberblocks, support for larger sector sizes, better rewind and diagnostics.
- AnyRaid-Z: Flexible RAID-Z with mixed disk sizes for higher usable capacity.
- Forced Export: Allows safe forced unmount of suspended pools to restore uptime.
- AWS Enhancements: Optimized write strategies for provisioned IOPS EBS volumes.
Potential Features
- BRT Log: Improves performance of block cloning/reflinks.
- SMR Drive Support: Optimizations to handle overlapping track designs.
Future Technologies
- NVMe-connected HDDs: Standardized interfaces for reconfigurable storage.
- CXL (Compute Express Link): Disaggregated computing and memory pooling to scale ZFS without distributed FS complexity.
https://klarasystems.com/articles/zfs-new-features-roadmap-innovations/
-
An initiative of Klara Inc to launch a Webinar with the most experienced devs in the ZFS storage industry.
-
Major features include:
- Quotas: Allow setting default user/group/project quotas (#17130)
- Uncached IO: Direct IO fallback to a light-weight uncached IO when unaligned (#17218)
- Unified allocation throttling: A new algorithm designed to reduce vdev fragmentation (#17020)
- Better encryption performance using AVX2 for AES-GCM (#17058)
- Allow ZIL on special vdevs when available (#17505)
- Extend
special_small_blocksto land ZVOL writes on special vdevs (#14876), and allow non-power of two values (#17497) - Add
zfs rewrite -Pwhich preserves logical birth time when possible to minimize incremental stream size (#17565) - Add
-a|--alloption which scrubs, trims, or initializes all imported pools (#17524) - Add
zpool scrub -S -Eto scrub specific time ranges (#16853) - Release topology restrictions on special/dedup vdevs (#17496)
- Multiple gang blocks improvements and fixes (#17111, #17004, #17587, #17484, #17123, #17073)
- New dedup optimizations and fixes (#17038 , #17123 , #17435, #17391)
-
This feature will allow the maximal use of storage in a ZFS pool consisted by drives of different capacity.
https://hexos.com/blog/introducing-zfs-anyraid-sponsored-by-eshtek
-
- Compatible with Linux 6.14
- Support for FreeBSD RELEASE >= 12.2
- Fixes for the GCC 15 compiler builds
-
“Key Features in OpenZFS 2.3.0:
RAIDZ Expansion (#15022): Add new devices to an existing RAIDZ pool, increasing storage capacity without downtime. Fast Dedup (#15896): A major performance upgrade to the original OpenZFS deduplication functionality. Direct IO (#10018): Allows bypassing the ARC for reads/writes, improving performance in scenarios like NVMe devices where caching may hinder efficiency. JSON (#16217): Optional JSON output for the most used commands. Long names (#15921): Support for file and directory names up to 1023 characters. Bug Fixes: A series of critical bug fixes addressing issues reported in previous versions. Numerous performance improvements throughout the code base. Supported Platforms: Linux kernels 4.18 - 6.12, FreeBSD releases 13.3, 14.0 - 14.2.1041 commits and 999 changed files
-
433 FreeBSD commits and 175 contributions to my favorite filesystem ZFS💪🏾
-
In Portland, Oregon.
-
“Performance delta focuses on the relationship of each dedup routine compared to un-deduped storage on the same hardware. It’s the same data but organized differently.
Fast dedup outperforms legacy dedup by almost 25% of the raw performance.“
/root 