[SPARK-21501] Spark shuffle index cache size should be memory based

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.1.0
• Fix Version/s: 2.3.0
• Component/s: Shuffle, Spark Core
• Assignee: Sanket Reddy
• Reporter: Thomas Graves
• Created: 21/Jul/17 16:21
• Updated: 21/Oct/20 13:11
• Resolved: 23/Aug/17 16:51

Description

Right now the spark shuffle service has a cache for index files. It is based on a # of files cached (spark.shuffle.service.index.cache.entries). This can cause issues if people have a lot of reducers because the size of each entry can fluctuate based on the # of reducers.

We saw an issues with a job that had 170000 reducers and it caused NM with spark shuffle service to use 700-800MB or memory in NM by itself.

We should change this cache to be memory based and only allow a certain memory size used. When I say memory based I mean the cache should have a limit of say 100MB.

Issue Links

• links to: https://github.com/apache/spark/pull/18940