Github user viirya commented on a diff in the pull request: https://github.com/apache/spark/pull/16989#discussion_r117606109 --- Diff: core/src/main/scala/org/apache/spark/internal/config/package.scala --- @@ -278,4 +278,21 @@ package object config { "spark.io.compression.codec.") .booleanConf .createWithDefault(false) + + private[spark] val SHUFFLE_ACCURATE_BLOCK_THRESHOLD = + ConfigBuilder("spark.shuffle.accurateBlkThreshold") + .doc("When we compress the size of shuffle blocks in HighlyCompressedMapStatus, we will " + + "record the size accurately if it's above the threshold specified by this config. This " + + "helps to prevent OOM by avoiding underestimating shuffle block size when fetch shuffle " + + "blocks.") + .longConf + .createWithDefault(100 * 1024 * 1024) + + private[spark] val REDUCER_MAX_REQ_SIZE_SHUFFLE_TO_MEM = + ConfigBuilder("spark.reducer.maxReqSizeShuffleToMem") --- End diff -- If we want to avoid fetch big requests into memory, shall we simply do this for all blocks sized over `SHUFFLE_ACCURATE_BLOCK_THRESHOLD`? Instead of going to compare with `REDUCER_MAX_REQ_SIZE_SHUFFLE_TO_MEM` at fetching time? To have two configs for the one final goal seems to cause more complexity.
--- 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 infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org