jackylk commented on a change in pull request #3637:
[CARBONDATA-3721][CARBONDATA-3590] Optimize Bucket Table
URL: https://github.com/apache/carbondata/pull/3637#discussion_r386769585
##########
File path:
integration/spark/src/main/scala/org/apache/carbondata/spark/rdd/CarbonMergerRDD.scala
##########
@@ -584,36 +591,53 @@ class CarbonMergerRDD[K, V](
logInfo("no.of.nodes where data present=" + nodeBlockMap.size())
defaultParallelism = sparkContext.defaultParallelism
- // Create Spark Partition for each task and assign blocks
- nodeBlockMap.asScala.foreach { case (nodeName, splitList) =>
- val taskSplitList = new java.util.ArrayList[NodeInfo](0)
- nodeTaskBlocksMap.put(nodeName, taskSplitList)
- var blockletCount = 0
- splitList.asScala.foreach { splitInfo =>
- val splitsPerNode = splitInfo.asInstanceOf[CarbonInputSplitTaskInfo]
- blockletCount = blockletCount +
splitsPerNode.getCarbonInputSplitList.size()
- taskSplitList.add(
- NodeInfo(splitsPerNode.getTaskId,
splitsPerNode.getCarbonInputSplitList.size()))
-
- if (blockletCount != 0) {
- val taskInfo = splitInfo.asInstanceOf[CarbonInputSplitTaskInfo]
- val multiBlockSplit = if (null == rangeColumn || singleRange) {
- new CarbonMultiBlockSplit(
- taskInfo.getCarbonInputSplitList,
- Array(nodeName))
- } else {
- var splitListForRange = new util.ArrayList[CarbonInputSplit]()
- new CarbonMultiBlockSplit(
- splitListForRange,
- Array(nodeName))
+ if (bucketInfo != null) {
Review comment:
extract these new logic into a private function, so not to make
internalCompute too big
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
