Github user viirya commented on a diff in the pull request:
https://github.com/apache/spark/pull/18031#discussion_r117610528
--- Diff: core/src/main/scala/org/apache/spark/scheduler/MapStatus.scala ---
@@ -193,8 +219,27 @@ private[spark] object HighlyCompressedMapStatus {
} else {
0
}
+ val threshold1 = Option(SparkEnv.get)
+ .map(_.conf.get(config.SHUFFLE_ACCURATE_BLOCK_THRESHOLD))
+ .getOrElse(config.SHUFFLE_ACCURATE_BLOCK_THRESHOLD.defaultValue.get)
+ val threshold2 = avgSize * Option(SparkEnv.get)
+
.map(_.conf.get(config.SHUFFLE_ACCURATE_BLOCK_THRESHOLD_BY_TIMES_AVERAGE))
+
.getOrElse(config.SHUFFLE_ACCURATE_BLOCK_THRESHOLD_BY_TIMES_AVERAGE.defaultValue.get)
+ val threshold = math.max(threshold1, threshold2)
+ val hugeBlockSizesArray = ArrayBuffer[Tuple2[Int, Byte]]()
+ if (numNonEmptyBlocks > 0) {
+ i = 0
+ while (i < totalNumBlocks) {
+ if (uncompressedSizes(i) > threshold) {
+ hugeBlockSizesArray += Tuple2(i,
MapStatus.compressSize(uncompressedSizes(i)))
+
+ }
+ i += 1
+ }
+ }
emptyBlocks.trim()
emptyBlocks.runOptimize()
- new HighlyCompressedMapStatus(loc, numNonEmptyBlocks, emptyBlocks,
avgSize)
+ new HighlyCompressedMapStatus(loc, numNonEmptyBlocks, emptyBlocks,
avgSize,
--- End diff --
In current change, if almost all blocks are huge, that's said it is not a
skew case, so we won't mark the blocks as huge ones. Then we will still fetch
them into memory?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
