xuanyuanking commented on a change in pull request #26040: [SPARK-9853][Core]
Optimize shuffle fetch of continuous partition IDs
URL: https://github.com/apache/spark/pull/26040#discussion_r332473606
##########
File path:
common/network-shuffle/src/main/java/org/apache/spark/network/shuffle/ExternalBlockHandler.java
##########
@@ -105,6 +105,11 @@ protected void handleMessage(
for (int[] ids: msg.reduceIds) {
numBlockIds += ids.length;
}
+ // For fetching continuous shuffle blocks in batch, we store both
start and end reduce
+ // id in reduceIds, so the real number of ShuffleBlockBatchId is
total length / 2.
Review comment:
Copy, done in 407b1e0.
----------------------------------------------------------------
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]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
