Github user andrewor14 commented on a diff in the pull request:
https://github.com/apache/spark/pull/1165#discussion_r14810095
--- Diff: core/src/main/scala/org/apache/spark/SparkEnv.scala ---
@@ -67,10 +67,14 @@ class SparkEnv (
val metricsSystem: MetricsSystem,
val conf: SparkConf) extends Logging {
- // A mapping of thread ID to amount of memory used for shuffle in bytes
+ // A mapping of thread ID to amount of memory, in bytes, used for shuffle
// All accesses should be manually synchronized
val shuffleMemoryMap = mutable.HashMap[Long, Long]()
+ // A mapping of thread ID to amount of memory, in bytes, used for
unrolling an RDD partition
+ // All accesses should be manually synchronized
+ val cacheMemoryMap = mutable.HashMap[Long, Long]()
--- End diff --
The name is out of date. This map used to be only for CacheManager, but now
it's for unrolling blocks in memory in general. So it should be something more
general than `partitionMemoryMap`...
---
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.
---
