Ngone51 commented on a change in pull request #31643:
URL: https://github.com/apache/spark/pull/31643#discussion_r585528211
##########
File path:
common/network-shuffle/src/main/java/org/apache/spark/network/shuffle/OneForOneBlockFetcher.java
##########
@@ -138,9 +138,24 @@ private FetchShuffleBlocks createFetchShuffleBlocksMsg(
}
long[] mapIds = Longs.toArray(mapIdToReduceIds.keySet());
int[][] reduceIdArr = new int[mapIds.length][];
+ int blockIdIndex = 0;
for (int i = 0; i < mapIds.length; i++) {
reduceIdArr[i] = Ints.toArray(mapIdToReduceIds.get(mapIds[i]));
+ // The `blockIds`'s order must be same with the read order specified in
in FetchShuffleBlocks
+ // because the shuffle data's return order should match the `blockIds`'s
order to ensure
+ // blockId and data match.
+ if (!batchFetchEnabled) {
+ for (int j = 0; j < reduceIdArr[i].length; j++) {
+ this.blockIds[blockIdIndex++] = "shuffle_" + shuffleId + "_" +
mapIds[i] + "_"
+ + reduceIdArr[i][j];
Review comment:
I think our original motivation was to avoid the reconstruction of
`blockId`, which avoids the intermediate object as well. And now we have
avoided the reconstruction but introduced another new intermediate object -
`BlocksInfo`. Does it become more efficient or more readable?
As for you think https://github.com/Ngone51/spark/pull/7/files is less
readable...I personally don't really think it's hard to read..perhaps adding
more comments should help.
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
