Github user YuhuWang2002 commented on the issue:
    following the review comment, I rewrite code for read a range of 
maps。like this:
    class BlockStoreShuffleReader[K, C](
        handle: BaseShuffleHandle[K, _, C],
        startPartition: Int,
        endPartition: Int,
        context: TaskContext,
        serializerManager: SerializerManager = SparkEnv.get.serializerManager,
        blockManager: BlockManager = SparkEnv.get.blockManager,
        mapOutputTracker: MapOutputTracker = SparkEnv.get.mapOutputTracker,
        startMapId: Option[Int] = None,
        endMapId: Option[Int] = None)
    To decide how many range for read from the maps。Use the 
spark.sql.adaptive.skewjoin.threshold value。We think the output size less 
than the skew threshold, It can handling in one task,else we split to many 
task,which every one task handing data size slightly less the skew threshold

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 or file a JIRA ticket
with INFRA.

To unsubscribe, e-mail:
For additional commands, e-mail:

Reply via email to