Github user mateiz commented on a diff in the pull request:
https://github.com/apache/spark/pull/50#discussion_r10201195
--- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala ---
@@ -71,10 +71,21 @@ private[spark] class CacheManager(blockManager:
BlockManager) extends Logging {
val computedValues = rdd.computeOrReadCheckpoint(split, context)
// Persist the result, so long as the task is not running locally
if (context.runningLocally) { return computedValues }
- val elements = new ArrayBuffer[Any]
- elements ++= computedValues
- blockManager.put(key, elements, storageLevel, tellMaster = true)
- elements.iterator.asInstanceOf[Iterator[T]]
+ if (storageLevel.useDisk && !storageLevel.useMemory) {
--- End diff --
This is not the only condition where we want to do this. For example we
might also want it for MEMORY_ONLY_SER, where the serialized data might fit in
RAM but the ArrayBuffer of raw objects might not. (Especially if you set
spark.rdd.compress to compress the serialized data.)
---
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.
---
