Wei-Chiu Chuang created HDDS-12518:
--------------------------------------
Summary: Compact RocksDB after consecutive deletions
Key: HDDS-12518
URL: https://issues.apache.org/jira/browse/HDDS-12518
Project: Apache Ozone
Issue Type: Improvement
Reporter: Wei-Chiu Chuang
Attachments: Screenshot 2025-03-07 at 6.15.03 PM.png, Screenshot
2025-03-07 at 6.15.10 PM.png, Screenshot 2025-03-07 at 6.15.49 PM.png
We can manually trigger compaction for RocksDB column families after some
deletions to improve seek performance and thus improve overall OM performance.
This is similar to RocksDB's builtin CompactOnDeleteCollector, but that
requires newer version of RocksDB so we can't use it yet.
I added a background service CompactionService to periodically (default 5
minutes) check if a column family has accumulated 10,000 deletes, and schedule
a full compaction if it does.
Using this heuristic, I was able to improve the deletion of a large directory
from 20k directories per hour to 400k per hour on my machine. The average
rocksdb seek time managed to maintain below 2ms. (Without compactions, average
db seek time would grow towards 20ms after several hours)
I also did a proof of concept to simply compact over and over again, in which
cause it could increase to 1 million deletions per hour. But that would be too
intense if there are foreground traffic at the same time. So a less frequent
approach is picked.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
