[CASSANDRA-3855] RemoveDeleted dominates compaction time for large sstable counts

• Type: Bug
• Status: Resolved
• Priority: Normal
• Resolution: Fixed
• Fix Version/s: 1.2.0 beta 1
• Labels: compaction deletes leveled
• Severity: Normal
• Assignee: Yuki Morishita
• Reporter: Stu Hood

Authors

Yuki Morishita

Reviewers

Jonathan Ellis

• Created: 05/Feb/12 21:52
• Updated: 16/Apr/19 09:32
• Resolved: 20/Jul/12 22:35

Description

With very large numbers of sstables (2000+ generated by a `bin/stress -n 100,000,000` run with LeveledCompactionStrategy), PrecompactedRow.removeDeletedAndOldShards dominates compaction runtime, such that commenting it out takes compaction throughput from 200KB/s to 12MB/s.

Stack attached.