waitinfuture commented on code in PR #2524:
URL: https://github.com/apache/celeborn/pull/2524#discussion_r1609441470
##########
worker/src/main/java/org/apache/celeborn/service/deploy/worker/storage/PartitionFilesSorter.java:
##########
@@ -629,10 +629,18 @@ public void sort() throws InterruptedException {
for (ShuffleBlockInfo blockInfo : originShuffleBlocks) {
long offset = blockInfo.offset;
long length = blockInfo.length;
- ShuffleBlockInfo sortedBlock = new ShuffleBlockInfo();
- sortedBlock.offset = fileIndex;
- sortedBlock.length = length;
- sortedShuffleBlocks.add(sortedBlock);
+ // combine multiple small length `ShuffleBlockInfo` for same mapId
such that
+ // size of compacted `ShuffleBlockInfo` does not exceed
`shuffleChunkSize`
+ if (!sortedShuffleBlocks.isEmpty()
+ && sortedShuffleBlocks.get(sortedShuffleBlocks.size() -
1).length + length
+ <= shuffleChunkSize) {
Review Comment:
Yes, say `shuffleChunkSize` is 8m, the generated blocks can be two 7.9m
chunks, when fetch data it can return the two chunks together, which is 15.8m.
That's the reason why I mentioned in the jira to use half of
`shuffleChunkSize`, in which the worst case is 3.9m * 3 = 11.7m. So multiply by
a fraction can control the exceeded size, maybe 0.25 is better (upper bound for
default config is 8m + 0.25 * 8m = 10m)
--
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]
