[ https://issues.apache.org/jira/browse/SPARK-21425?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16095323#comment-16095323 ]
Shixiong Zhu commented on SPARK-21425: -------------------------------------- [~srowen] The issue is static accumulators. Right? They won't be serialized/deserialized with tasks and cannot be reported back to driver. If running in local-cluster mode or a real cluster, they will always be 0 in the driver side. The overhead comes from memory barrier introduced by synchronization. I tried to improved this but gave up due to the significant performance regression: https://github.com/apache/spark/pull/15065 > LongAccumulator, DoubleAccumulator not threadsafe > ------------------------------------------------- > > Key: SPARK-21425 > URL: https://issues.apache.org/jira/browse/SPARK-21425 > Project: Spark > Issue Type: Bug > Components: Spark Core > Affects Versions: 2.2.0 > Reporter: Ryan Williams > Priority: Minor > > [AccumulatorV2 > docs|https://github.com/apache/spark/blob/v2.2.0/core/src/main/scala/org/apache/spark/util/AccumulatorV2.scala#L42-L43] > acknowledge that accumulators must be concurrent-read-safe, but afaict they > must also be concurrent-write-safe. > The same docs imply that {{Int}} and {{Long}} meet either/both of these > criteria, when afaict they do not. > Relatedly, the provided > [LongAccumulator|https://github.com/apache/spark/blob/v2.2.0/core/src/main/scala/org/apache/spark/util/AccumulatorV2.scala#L291] > and > [DoubleAccumulator|https://github.com/apache/spark/blob/v2.2.0/core/src/main/scala/org/apache/spark/util/AccumulatorV2.scala#L370] > are not thread-safe, and should be expected to behave undefinedly when > multiple concurrent tasks on the same executor write to them. > [Here is a repro repo|https://github.com/ryan-williams/spark-bugs/tree/accum] > with some simple applications that demonstrate incorrect results from > {{LongAccumulator}}'s. -- This message was sent by Atlassian JIRA (v6.4.14#64029) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org