[ https://issues.apache.org/jira/browse/DRILL-4152?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15056917#comment-15056917 ]
ASF GitHub Bot commented on DRILL-4152: --------------------------------------- Github user adeneche commented on a diff in the pull request: https://github.com/apache/drill/pull/298#discussion_r47573483 --- Diff: exec/java-exec/src/main/java/org/apache/drill/exec/store/parquet/columnreaders/ParquetRecordReader.java --- @@ -105,13 +107,15 @@ long totalRecordsRead; private final FragmentContext fragmentContext; + public ParquetReaderStats parquetReaderStats = new ParquetReaderStats(); --- End diff -- should we mark this as `final` ? it makes it easier for other developers understand how it's supposed to be used. Of course, we'll have to remove the `set to null` in the `close()` method but I guess that's fine unless of course the close method can be called multiple times > Add additional logging and metrics to the Parquet reader > -------------------------------------------------------- > > Key: DRILL-4152 > URL: https://issues.apache.org/jira/browse/DRILL-4152 > Project: Apache Drill > Issue Type: Bug > Components: Storage - Parquet > Reporter: Parth Chandra > Assignee: Deneche A. Hakim > > In some cases, we see the Parquet reader as the bottleneck in reading from > the file system. RWSpeedTest is able to read 10x faster than the Parquet > reader so reading from disk is not the issue. This issue is to add more > instrumentation to the Parquet reader so speed bottlenecks can be better > diagnosed. -- This message was sent by Atlassian JIRA (v6.3.4#6332)