[SPARK-1112] When spark.akka.frameSize > 10, task results bigger than 10MiB block execution

• Type: Bug
• Status: Resolved
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 0.9.0, 1.0.0
• Fix Version/s: 0.9.2, 1.0.1, 1.1.0
• Component/s: Spark Core
• Target Version/s: 0.9.2, 1.0.1, 1.1.0
• Assignee: Xiangrui Meng
• Reporter: Guillaume Pitel
• Created: 20/Feb/14 08:34
• Updated: 01/Dec/14 08:29
• Resolved: 17/Jul/14 04:31

Description

When I set the spark.akka.frameSize to something over 10, the messages sent from the executors to the driver completely block the execution if the message is bigger than 10MiB and smaller than the frameSize (if it's above the frameSize, it's ok)

Workaround is to set the spark.akka.frameSize to 10. In this case, since 0.8.1, the blockManager deal with the data to be sent. It seems slower than akka direct message though.

The configuration seems to be correctly read (see actorSystemConfig.txt), so I don't see where the 10MiB could come from

Issue Links

• is related to: SPARK-2138 The KMeans algorithm in the MLlib can lead to the Serialized Task size become bigger and bigger