[
https://issues.apache.org/jira/browse/HUDI-2161?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17378547#comment-17378547
]
ASF GitHub Bot commented on HUDI-2161:
--------------------------------------
nsivabalan commented on a change in pull request #3247:
URL: https://github.com/apache/hudi/pull/3247#discussion_r666681609
##########
File path:
hudi-spark-datasource/hudi-spark/src/main/java/org/apache/hudi/HoodieDatasetBulkInsertHelper.java
##########
@@ -110,6 +151,55 @@
Dataset<Row> colOrderedDataset = dedupedDf.select(
JavaConverters.collectionAsScalaIterableConverter(orderedFields).asScala().toSeq());
- return bulkInsertPartitionerRows.repartitionRecords(colOrderedDataset,
config.getBulkInsertShuffleParallelism());
+ return Pair.of(populateMetaCols ?
bulkInsertPartitionerRows.repartitionRecords(colOrderedDataset,
config.getBulkInsertShuffleParallelism()) :
+ new NonSortPartitionerWithRows().repartitionRecords(colOrderedDataset,
config.getBulkInsertShuffleParallelism()), nonPartitionedDataset);
}
+
+ /**
+ * Add empty meta columns and reorder such that meta columns are at the
beginning.
+ *
+ * @param rows
+ * @return
+ */
+ public static Dataset<Row>
prepareHoodieDatasetForBulkInsertAppendOnly(Dataset<Row> rows,
HoodieWriteConfig config, boolean isGlobalIndex) {
Review comment:
this method is not used. both flows use the previous method.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
> Add support to disable meta column to BulkInsert Row Writer path
> ----------------------------------------------------------------
>
> Key: HUDI-2161
> URL: https://issues.apache.org/jira/browse/HUDI-2161
> Project: Apache Hudi
> Issue Type: Improvement
> Reporter: sivabalan narayanan
> Priority: Major
>
> Objective here is to disable all meta columns so as to avoid storage cost.
> Also, some benefits could be seen in write latency with row writer path as no
> special handling is required at RowCreateHandle layer.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
