You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ignite.apache.org by "Konstantin Orlov (Jira)" <ji...@apache.org> on 2026/04/07 14:39:00 UTC

[jira] [Created] (IGNITE-28483) Sql. Simplify close method in DistributedQueryManager

Konstantin Orlov created IGNITE-28483:
-----------------------------------------

             Summary: Sql. Simplify close method in DistributedQueryManager
                 Key: IGNITE-28483
                 URL: https://issues.apache.org/jira/browse/IGNITE-28483
             Project: Ignite
          Issue Type: Improvement
          Components: sql ai3
            Reporter: Konstantin Orlov
            Assignee: Konstantin Orlov


Currently, this method first build execution chain, and then starts execution in a separate thread pool. But synchronous part is in fact really small: it basically schedules the closing task for every local fragment.

The problem is that this method is on hot path, and introduces one unnecessary thread switch, which in turn may affect latency in certain scenarios.

Second problem with this method is that it sends `queryCloseMessage` unconditionally back to coordinator node. Need to take closer look and probably get rid of this.



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