You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ignite.apache.org by "Vladimir Ozerov (JIRA)" <ji...@apache.org> on 2018/12/06 08:51:00 UTC

[jira] [Created] (IGNITE-10566) SQL: Map query should operate on it's own list of partitions

Vladimir Ozerov created IGNITE-10566:
----------------------------------------

             Summary: SQL: Map query should operate on it's own list of partitions
                 Key: IGNITE-10566
                 URL: https://issues.apache.org/jira/browse/IGNITE-10566
             Project: Ignite
          Issue Type: Task
          Components: sql
            Reporter: Vladimir Ozerov


Currently we merge extracted partitions from multiple map queries using {{OR}} semantics. The this list is used for execution of all queries. That is, consider that user's query was split into two map queries {{A}} and {{B}} with target partitions {{1}} and {{2}} respectively. Instead of executing {{A -> 1}} and {{B- > 2}} we will execute {{A -> 1, 2}} and {{B -> 1, 2}}, what leads to waste of resources. 

See {{GridH2QueryRequest#qryParts}} as a starting point.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)