[
https://issues.apache.org/jira/browse/DRILL-7199?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16831171#comment-16831171
]
ASF GitHub Bot commented on DRILL-7199:
---------------------------------------
dvjyothsna commented on pull request #1771: DRILL-7199: Optimize population of
metadata for non-interesting columns
URL: https://github.com/apache/drill/pull/1771#discussion_r280169434
##########
File path:
exec/java-exec/src/main/java/org/apache/drill/exec/physical/base/AbstractGroupScanWithMetadata.java
##########
@@ -146,15 +148,23 @@ public boolean isMatchAllMetadata() {
*/
@Override
public long getColumnValueCount(SchemaPath column) {
- long tableRowCount = (long)
TableStatisticsKind.ROW_COUNT.getValue(getTableMetadata());
+ long tableRowCount, colNulls;
+ Long nulls;
ColumnStatistics columnStats =
getTableMetadata().getColumnStatistics(column);
- long colNulls;
+ ColumnStatistics nonInterestingColStats =
getNonInterestingColumnsMetadata().getColumnStatistics(column);
+
if (columnStats != null) {
- Long nulls = (Long)
columnStats.getStatistic(ColumnStatisticsKind.NULLS_COUNT);
- colNulls = nulls != null ? nulls : GroupScan.NO_COLUMN_STATS;
+ tableRowCount = (long)
TableStatisticsKind.ROW_COUNT.getValue(getTableMetadata());
+ } else if (nonInterestingColStats != null) {
Review comment:
Yes, tablemetadata holds only the interesting column stats. ColumnStats will
be non-null only for the interesting columns. There are three scenarios here:
- Column is interesting. Then the columnStats is not null and it has the
correct row count.
- Column is non-interesting. Then the columnStats will be null and
nonInterestingColStats is not null and the row count is marked as unknown.
- Column doesn't exist in the table. So both columnStats and
nonInterestingColStats will be null. Since the column doesn't exist, the row
count is set to 0.
Here we know that a column will only belong to one of the above cases. Thats
the reason there is an else-if instead of if.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Optimize the time taken to populate column statistics for non-interesting
> columns
> ---------------------------------------------------------------------------------
>
> Key: DRILL-7199
> URL: https://issues.apache.org/jira/browse/DRILL-7199
> Project: Apache Drill
> Issue Type: Bug
> Reporter: Venkata Jyothsna Donapati
> Assignee: Venkata Jyothsna Donapati
> Priority: Minor
> Fix For: 1.17.0
>
> Original Estimate: 48h
> Remaining Estimate: 48h
>
> Currently populating column statistics for non-interesting columns very long
> since it is populated for every row group. Since non-interesting column
> statistics are common for the table, it can be populated once and can be
> reused.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
