[
https://issues.apache.org/jira/browse/HIVE-25827?focusedWorklogId=781631&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-781631
]
ASF GitHub Bot logged work on HIVE-25827:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 15/Jun/22 12:56
Start Date: 15/Jun/22 12:56
Worklog Time Spent: 10m
Work Description: pvary commented on code in PR #3368:
URL: https://github.com/apache/hive/pull/3368#discussion_r897944789
##########
ql/src/java/org/apache/hadoop/hive/ql/io/parquet/vector/VectorizedParquetRecordReader.java:
##########
@@ -138,32 +130,53 @@ public class VectorizedParquetRecordReader extends
ParquetRecordReaderBase
private ZoneId writerTimezone;
private final BucketIdentifier bucketIdentifier;
- public VectorizedParquetRecordReader(
- org.apache.hadoop.mapred.InputSplit oldInputSplit, JobConf conf) {
+ // LLAP cache integration
+ // TODO: also support fileKey in splits, like OrcSplit does
+ private Object cacheKey = null;
+ private CacheTag cacheTag = null;
+
+ public VectorizedParquetRecordReader(InputSplit oldInputSplit, JobConf conf)
throws IOException {
this(oldInputSplit, conf, null, null, null);
}
public VectorizedParquetRecordReader(
- org.apache.hadoop.mapred.InputSplit oldInputSplit, JobConf conf,
- FileMetadataCache metadataCache, DataCache dataCache, Configuration
cacheConf) {
+ InputSplit oldInputSplit, JobConf conf,
+ FileMetadataCache metadataCache, DataCache dataCache, Configuration
cacheConf)
+ throws IOException {
+ super(conf, oldInputSplit);
try {
this.metadataCache = metadataCache;
this.cache = dataCache;
this.cacheConf = cacheConf;
- serDeStats = new SerDeStats();
- projectionPusher = new ProjectionPusher();
+
+ if (metadataCache != null) {
+ cacheKey = HdfsUtils.getFileId(filePath.getFileSystem(conf), filePath,
Review Comment:
Do we have this code somewhere on the ORC codepath as well?
Issue Time Tracking
-------------------
Worklog Id: (was: 781631)
Time Spent: 1h (was: 50m)
> Parquet file footer is read multiple times, when multiple splits are created
> in same file
> -----------------------------------------------------------------------------------------
>
> Key: HIVE-25827
> URL: https://issues.apache.org/jira/browse/HIVE-25827
> Project: Hive
> Issue Type: Improvement
> Reporter: Rajesh Balamohan
> Assignee: Ádám Szita
> Priority: Major
> Labels: performance, pull-request-available
> Attachments: image-2021-12-21-03-19-38-577.png
>
> Time Spent: 1h
> Remaining Estimate: 0h
>
> With large files, it is possible that multiple splits are created in the same
> file. With current codebase, "ParquetRecordReaderBase" ends up reading file
> footer for each split.
> It can be optimized not to read footer information multiple times for the
> same file.
>
> [https://github.com/apache/hive/blob/master/ql/src/java/org/apache/hadoop/hive/ql/io/parquet/vector/VectorizedParquetRecordReader.java#L160]
>
> [https://github.com/apache/hive/blob/master/ql/src/java/org/apache/hadoop/hive/ql/io/parquet/ParquetRecordReaderBase.java#L91]
>
>
> !image-2021-12-21-03-19-38-577.png|width=1363,height=1256!
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
