[jira] [Commented] (PHOENIX-2724) Query with large number of guideposts is slower compared to no stats

["James Taylor (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/PHOENIX-2724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15249180#comment-15249180 ] 

James Taylor commented on PHOENIX-2724:
---------------------------------------

+1 - nice work! I don't think you need this, as the parent BaseResultIterators won't call through to it. You should be able to verify through the QueryPlanTest.
{code}
+        @Override
+        public void explain(List<String> planSteps) {
+            //TODO: samarth verify if this is the right thing to do here.
+            try {
+                currentIterator().explain(planSteps);
+            } catch (SQLException e) {
+                throw new RuntimeException(e);
+            }
+        }
{code}

> Query with large number of guideposts is slower compared to no stats
> --------------------------------------------------------------------
>
>                 Key: PHOENIX-2724
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-2724
>             Project: Phoenix
>          Issue Type: Bug
>    Affects Versions: 4.7.0
>         Environment: Phoenix 4.7.0-RC4, HBase-0.98.17 on a 8 node cluster
>            Reporter: Mujtaba Chohan
>            Assignee: Samarth Jain
>             Fix For: 4.8.0
>
>         Attachments: PHOENIX-2724.patch
>
>
> With 1MB guidepost width for ~900GB/500M rows table. Queries with short scan range gets significantly slower.
> Without stats:
> {code}
> select * from T limit 10; // query execution time <100 msec
> {code}
> With stats:
> {code}
> select * from T limit 10; // query execution time >20 seconds
> Explain plan: CLIENT 876085-CHUNK 476569382 ROWS 876060986727 BYTES SERIAL 1-WAY FULL SCAN OVER T SERVER 10 ROW LIMIT CLIENT 10 ROW LIMIT
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)