Github user pwendell commented on a diff in the pull request:
https://github.com/apache/spark/pull/4050#discussion_r23131557
--- Diff: core/src/main/scala/org/apache/spark/rdd/HadoopRDD.scala ---
@@ -219,6 +220,9 @@ class HadoopRDD[K, V](
val bytesReadCallback = if
(split.inputSplit.value.isInstanceOf[FileSplit]) {
SparkHadoopUtil.get.getFSBytesReadOnThreadCallback(
split.inputSplit.value.asInstanceOf[FileSplit].getPath, jobConf)
+ } else if (split.inputSplit.value.isInstanceOf[CombineFileSplit]) {
+ SparkHadoopUtil.get.getFSBytesReadOnThreadCallback(
--- End diff --
Is it guaranteed that all paths in the `CombineFileSplit` have the same
filesystem?
Also one related question after I dug around a bit more. Hadoops
`FileSystem.getAllStatistics()` returns a list where you can only distinguish
one file system from another via the scheme. What happens if two different
`hdfs://` filesystems are being read from within the same thread (for instance
if two Hadoop RDD's are coalesced)? Is the assumption that this will never
happen?
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
