Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/8844#discussion_r40025584
--- Diff: core/src/main/scala/org/apache/spark/MapOutputTracker.scala ---
@@ -134,11 +134,24 @@ private[spark] abstract class MapOutputTracker(conf:
SparkConf) extends Logging
*/
def getMapSizesByExecutorId(shuffleId: Int, reduceId: Int)
: Seq[(BlockManagerId, Seq[(BlockId, Long)])] = {
- logDebug(s"Fetching outputs for shuffle $shuffleId, reduce $reduceId")
+ getMapSizesByExecutorId(shuffleId, reduceId, reduceId + 1)
+ }
+
+ /**
+ * Called from executors to get the server URIs and output sizes for
each shuffle block that
+ * needs to be read from a given range of map output partitions.
+ *
+ * @return A sequence of 2-item tuples, where the first item in the
tuple is a BlockManagerId,
+ * and the second item is a sequence of (shuffle block id,
shuffle block size) tuples
+ * describing the shuffle blocks that are stored at that block
manager.
+ */
+ def getMapSizesByExecutorId(shuffleId: Int, startPartition: Int,
endPartition: Int)
--- End diff --
The Scaladoc here should probably explain whether the range includes the
`endPartition` endpoint.
---
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]
