[ https://issues.apache.org/jira/browse/DERBY-6938?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16088752#comment-16088752 ]
Bryan Pendleton commented on DERBY-6938: ---------------------------------------- Hi Vardhan, I think your observation about the Optimizer being unwilling to choose Hash Join plans if it believes they will not fit in memory is very interesting. Our documentation (https://db.apache.org/derby/docs/10.13/tuning/ctunoptimz23173.html) definitely claims that Derby will choose a hash join, and will spill to disk if the join is larger than will fit in memory, and I believe that was definitely the intent of DERBY-106. But it appears that it isn't working as we expect. In researching this, I came across a very interesting discussion in DERBY-4620. Perhaps you could look through the DERBY-4620 work, and see if you have any additional observations to share regarding the behavior that you see? > Obtain cardinality estimates and true estimates for base tables as well as > for intermediate results for queries involving multiple joins. > ------------------------------------------------------------------------------------------------------------------------------------------- > > Key: DERBY-6938 > URL: https://issues.apache.org/jira/browse/DERBY-6938 > Project: Derby > Issue Type: Sub-task > Components: SQL > Reporter: Harshvardhan Gupta > Assignee: Harshvardhan Gupta > Attachments: explain.txt > > -- This message was sent by Atlassian JIRA (v6.4.14#64029)