ajantha-bhat commented on a change in pull request #3538: [CARBONDATA-3637]
Optimize insert into flow
URL: https://github.com/apache/carbondata/pull/3538#discussion_r376343994
##########
File path:
integration/spark2/src/main/scala/org/apache/carbondata/spark/rdd/CarbonDataRDDFactory.scala
##########
@@ -957,24 +987,36 @@ object CarbonDataRDDFactory {
private def loadDataFrame(
sqlContext: SQLContext,
dataFrame: Option[DataFrame],
+ scanResultRDD: Option[RDD[InternalRow]],
carbonLoadModel: CarbonLoadModel
): Array[(String, (LoadMetadataDetails, ExecutionErrors))] = {
try {
- val rdd = dataFrame.get.rdd
-
+ val rdd = if (dataFrame.isDefined) {
+ dataFrame.get.rdd
+ } else {
+ scanResultRDD.get
+ }
val nodeNumOfData = rdd.partitions.flatMap[String, Array[String]] { p =>
DataLoadPartitionCoalescer.getPreferredLocs(rdd, p).map(_.host)
}.distinct.length
val nodes = DistributionUtil.ensureExecutorsByNumberAndGetNodeList(
nodeNumOfData,
sqlContext.sparkContext)
- val newRdd = new DataLoadCoalescedRDD[Row](sqlContext.sparkSession, rdd,
nodes.toArray
- .distinct)
-
+ val newRdd =
+ if (dataFrame.isDefined) {
+ new DataLoadCoalescedRDD[Row](sqlContext.sparkSession,
dataFrame.get.rdd, nodes.toArray
+ .distinct)
+ } else {
+ new DataLoadCoalescedRDD[InternalRow](sqlContext.sparkSession,
+ scanResultRDD.get,
+ nodes.toArray
+ .distinct)
Review comment:
yes, this flow is for local sort and no sort. Have plan to send no sort to
global sort plan.
local sort, one task one node. Hence this logic
----------------------------------------------------------------
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
