wget -O hdxctl https://hdx-infrastructure.s3.amazonaws.com/hdxctl-v2.14.4 && chmod +x hdxctl
- Add dedicated cache volume for query peer
- Default merge head period decreased to 1 minute from 5 minutes
- New table settings defaults to improve performance
- Refactor to support pod instances in client stack
- Latest version of Superset required
- Patched latest clickhouse driver for sqlalchemy
- Moved some instances to gp3 volumes (from gp2)
- Config API update to move shard_key to table settings
- Spot fleet capabilities added to the cluster.
- Significant stability fixes for long running, extreme data capacity queries.
- Increase proxy timeout between superset and heads
- Sort Keys Reworked for Merge and Write SDK
- Prometheus - persistent data across cluster replacement
- Fixed use of Pools.
- Bugfix for sort_keys and column sorting
- Multi column performance optimizations for query performance.
- HDXCTL bug updates.