[SPARK-24143] filter empty blocks when convert mapstatus to (blockId, size) pair

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.3.0
• Fix Version/s: 2.4.0
• Component/s: Spark Core
• Assignee: Jin Xing
• Reporter: Jin Xing
• Created: 02/May/18 05:39
• Updated: 07/May/18 06:19
• Resolved: 07/May/18 06:19

Description

In current code(MapOutputTracker.convertMapStatuses), mapstatus are converted to (blockId, size) pair for all blocks – no matter the block is empty or not, which result in OOM when there are lots of consecutive empty blocks, especially when adaptive execution is enabled.

(blockId, size) pair is only used in ShuffleBlockFetcherIterator to control shuffle-read and only non-empty block request is sent. Can we just filter out the empty blocks in MapOutputTracker.convertMapStatuses and save memory?

Issue Links

• links to: [Github] Pull Request #21212 (jinxing64)