[jira] [Updated] (HDDS-15991) Optimize QuotaRepairTask deleted table deserialization with batching

["ASF GitHub Bot (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HDDS-15991?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

ASF GitHub Bot updated HDDS-15991:
----------------------------------
    Labels: pull-request-available  (was: )

> Optimize QuotaRepairTask deleted table deserialization with batching
> --------------------------------------------------------------------
>
>                 Key: HDDS-15991
>                 URL: https://issues.apache.org/jira/browse/HDDS-15991
>             Project: Apache Ozone
>          Issue Type: Sub-task
>            Reporter: Chung-En Lee
>            Assignee: Chi-Hsuan Huang
>            Priority: Major
>              Labels: pull-request-available
>
> QuotaRepairTask uses a single thread when scanning deletedTable and deletedDirTable. When the active DB undergoes large-scale deletions, deserialization becomes a performance bottleneck and slows down the repair process.
> Solution:
>  # Follow the approach in recalculateUsages by converting the active DB's deletedTable / deletedDirTable scans to multi-threaded batch processing.
>  # Note: Scans for Snapshot DBs should remain single-threaded (or avoid sharing the same Executor pool) to prevent thread pool starvation caused by a large number of snapshots.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org