You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@kafka.apache.org by "Chia-Ping Tsai (Jira)" <ji...@apache.org> on 2026/05/17 08:14:00 UTC

[jira] [Created] (KAFKA-20595) Parallelize loadSegmentFiles

Chia-Ping Tsai created KAFKA-20595:
--------------------------------------

             Summary: Parallelize loadSegmentFiles
                 Key: KAFKA-20595
                 URL: https://issues.apache.org/jira/browse/KAFKA-20595
             Project: Kafka
          Issue Type: Improvement
            Reporter: Chia-Ping Tsai
            Assignee: Chia-Ping Tsai


This is related to KAFKA-20567. 

In the happy path where all segment files are intact and index rebuilding is not required, we can parallelize the log segment loading process. If we do encounter any corrupted or orphaned files during the parallel scan, we can collect them into a list and handle the recovery sequentially afterward to ensure correctness



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