Github user aarondav commented on a diff in the pull request:
https://github.com/apache/spark/pull/158#discussion_r11227619
--- Diff: core/src/main/scala/org/apache/spark/storage/BlockManager.scala
---
@@ -620,6 +651,24 @@ private[spark] class BlockManager(
}
// Keep track of which blocks are dropped from memory
res.droppedBlocks.foreach { block => updatedBlocks += block }
+ } else if (level.useOffHeap) {
+ // Save to Tachyon.
+ val askForBytes = level.replication > 1
--- End diff --
According to Matei's comments, we don't need to support level.replication >
1, so we can eliminate this logic as well as the path in tachyonStore that
handles askForBytes = true. Woo!
---
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.
---
