[ https://issues.apache.org/jira/browse/SOLR-12572?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16553333#comment-16553333 ]
Amrit Sarkar commented on SOLR-12572: ------------------------------------- Attached WIP patch, which reuses docValues for "numeric" field types. Initial tests showed improvement up-to 30% in execution time against export handler, for 2.5M docs and sorting on 8 fields. I am facing IndexOutOfBounds exception for string and boolean (implements string logic) data type, as "exact" values are not computed, but "ordinals". I will detail this out more after performing few tests. > Reuse fieldvalues computed while sorting at writing in ExportWriter > ------------------------------------------------------------------- > > Key: SOLR-12572 > URL: https://issues.apache.org/jira/browse/SOLR-12572 > Project: Solr > Issue Type: Improvement > Security Level: Public(Default Security Level. Issues are Public) > Components: streaming expressions > Reporter: Amrit Sarkar > Assignee: Varun Thacker > Priority: Minor > Attachments: SOLR-12572.patch, SOLR-12572.patch > > > Today while exporting result through "/export" handler, > {code:java} > http://localhost:8983/solr/core_name/export?q=my-query&sort=severity+desc,timestamp+desc&fl=severity,timestamp,msg > {code} > Doc-values are sought for all the {{sort}} fields defined (in this example > 'severity, 'timestamp'). When we stream out docs we again make doc-value > seeks against the {{fl}} fields ('severity','timestamp','msg') . > In most common use-cases we have {{fl = sort}} fields, or atleast the sort > fields are subset of {{fl}} fields, so if we can *pre-collect* the values > while sorting it, we can reduce the doc-value seeks potentially bringing > *speed improvement*. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org