This is an UNOFFICIAL & tentative list of big new features expected in the glusterfs 3.3 series.
- Unified File & Object access.
- Hadoop hooks - HDFS compatibility layer
New Volume Type
- Repstr - replicated + striped (+ distributed) volumes
Enhancements to Distributed volumes (DHT translator)
- Rebalance can migrate open files
- Remove-brick can migrate data to remaining bricks
Enhancements to Replicated volumes (AFR translator)
- Ability to change replica count on an active volume (possibly including ability to add replication to distribute-only volumes)
- Granular locking - useful for running VM images needing to be healed. Glusterfs will not have to lock & freeze the whole VM disk image, only the parts that need healing.
- Proactive self-heal - when a brick comes back online the gluster daemon should begin the self-heal process without waiting for the files to be accessed.
- Round-trip reduction - should improve performance especially for smaller files.
- Quorum enforcement - should help resolve or avoid split brain scenarios
Thanks to jdarcy, kkeithley & bulde in #gluster for helping semiosis with the info for this.