[GitHub] spark pull request #21474: [SPARK-24297][CORE] Fetch-to-disk by default for ...

[felixcheung <gi...@git.apache.org>]

Github user felixcheung commented on a diff in the pull request:

    https://github.com/apache/spark/pull/21474#discussion_r192295318
  
    --- Diff: core/src/main/scala/org/apache/spark/internal/config/package.scala ---
    @@ -429,7 +429,11 @@ package object config {
             "external shuffle service, this feature can only be worked when external shuffle" +
             "service is newer than Spark 2.2.")
           .bytesConf(ByteUnit.BYTE)
    -      .createWithDefault(Long.MaxValue)
    +      // fetch-to-mem is guaranteed to fail if the message is bigger than 2 GB, so we might
    +      // as well use fetch-to-disk in that case.  The message includes some metadata in addition
    +      // to the block data itself (in particular UploadBlock has a lot of metadata), so we leave
    +      // extra room.
    +      .createWithDefault(Int.MaxValue - 500)
    --- End diff --
    
    is the fixed buffer of 500 bytes enough for metadata?


---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org