[REMOTE-SHUFFLE-43] Read Partitions from Highly Compressed Map Status #49
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: jiafu zhang <jiafu.zhang@intel.com>
Signed-off-by: jiafu zhang <jiafu.zhang@intel.com>
Signed-off-by: jiafu zhang <jiafu.zhang@intel.com>
… Driver Signed-off-by: jiafu zhang <jiafu.zhang@intel.com>
… Driver Signed-off-by: jiafu zhang <jiafu.zhang@intel.com>
Signed-off-by: jiafu zhang <jiafu.zhang@intel.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
The partition lengths in highly compressed map status is not accurate. We depended on the lengths to read data from partitions. It caused only partial data being read.
To fix the issue, we depend on the actual size returned from DAOS, if it's highly compressed map status, to judge if end of partition is reached. To maintain the perf, we read multiple partitions in parallel as much as possible whilst weigh on current partition so that reduce task experiences less I/O wait.
How was this patch tested?
It's tested with HiBench repartition. By adjusting the value of "spark.shuffle.minNumPartitionsToHighlyCompress", we can control if we use normal compressed map status or highly compressed map status.