You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Doug Rohrer (Jira)" <ji...@apache.org> on 2025/04/18 20:48:00 UTC

[jira] [Created] (CASSANALYTICS-39) Improved Compaction/Repair load for bulk writes

Doug Rohrer created CASSANALYTICS-39:
----------------------------------------

             Summary: Improved Compaction/Repair load for bulk writes
                 Key: CASSANALYTICS-39
                 URL: https://issues.apache.org/jira/browse/CASSANALYTICS-39
             Project: Apache Cassandra Analytics
          Issue Type: Task
            Reporter: Doug Rohrer


Today, the bulk writer imports unrepaired SSTables into level 0, which causes a significant amount of I/O during compaction and repair that could be avoided if we could coordinate the import somehow. This would allow us to leverage Cassandra’s knowledge of the current state of SSTables on-disk and the fact that the Bulk Writer never produces overlapping SSTables to import them into a higher level and, if all imports succeed for all replicas of a particular range, mark them as repaired.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org