Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

chore: Override node name for CometSparkToColumnar #958

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

chore: Override node name for CometSparkToColumnar #958

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
76c8952
Add method nodeName()
Sep 21, 2024
6b5ca95
Add unit test for row input
Sep 23, 2024
e0849d6
Add unit test for columnar input
JensonChoi Nov 8, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions spark/src/main/scala/org/apache/spark/sql/comet/CometSparkToColumnarExec.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,12 @@ case class CometSparkToColumnarExec(child: SparkPlan)

override def supportsColumnar: Boolean = true

override def nodeName: String = if (child.supportsColumnar) {
"CometSparkColumnarToColumnar"
} else {
"CometSparkRowToColumnar"
}

override lazy val metrics: Map[String, SQLMetric] = Map(
"numInputRows" -> SQLMetrics.createMetric(sparkContext, "number of input rows"),
"numOutputBatches" -> SQLMetrics.createMetric(sparkContext, "number of output batches"),
Expand Down
53 changes: 53 additions & 0 deletions spark/src/test/scala/org/apache/comet/exec/CometExecSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1707,6 +1707,59 @@ class CometExecSuite extends CometTestBase {
}
}

test("SparkToColumnar override node name for row input") {
withSQLConf(
SQLConf.ADAPTIVE_EXECUTION_ENABLED.key -> "true",
CometConf.COMET_SHUFFLE_MODE.key -> "jvm") {
val df = spark
.range(1000)
.selectExpr("id as key", "id % 8 as value")
.toDF("key", "value")
.groupBy("key")
.count()
df.collect()

val planAfter = df.queryExecution.executedPlan
assert(planAfter.toString.startsWith("AdaptiveSparkPlan isFinalPlan=true"))
val adaptivePlan = planAfter.asInstanceOf[AdaptiveSparkPlanExec].executedPlan
val nodeNames = adaptivePlan.collect { case c: CometSparkToColumnarExec =>
c.nodeName
}
assert(nodeNames.length == 1)
assert(nodeNames.head == "CometSparkRowToColumnar")
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you also add a test that will generate a plan that uses CometSparkColumnarToColumnar so that we are testing both cases?

I think you could have a copy of this test that writes the dataframe to a Parquet file and then reads the Parquet file back with the following configs. This will use Spark's vectorized Parquet reader which returns Spark columns.

        SQLConf.USE_V1_SOURCE_LIST.key -> "",
        CometConf.COMET_NATIVE_SCAN_ENABLED.key -> "false",
        CometConf.COMET_CONVERT_FROM_PARQUET_ENABLED.key -> "true") {

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hey @andygrove, I'm a bit stuck on the unit test for CometSparkColumnarToColumnar. I pushed a commit that contains what I've been working on, so you can take a look. However, I'm getting this error when I run the unit test:

  Cause: java.lang.ClassCastException: class org.apache.spark.sql.vectorized.ColumnarBatch cannot be cast to class org.apache.spark.sql.catalyst.InternalRow (org.apache.spark.sql.vectorized.ColumnarBatch and org.apache.spark.sql.catalyst.InternalRow are in unnamed module of loader 'app')
  at scala.collection.Iterator$$anon$10.next(Iterator.scala:461)
  at org.apache.spark.sql.execution.SparkPlan.$anonfun$getByteArrayRdd$1(SparkPlan.scala:389)
  at org.apache.spark.rdd.RDD.$anonfun$mapPartitionsInternal$2(RDD.scala:891)
  at org.apache.spark.rdd.RDD.$anonfun$mapPartitionsInternal$2$adapted(RDD.scala:891)
  at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
  at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:367)
  at org.apache.spark.rdd.RDD.iterator(RDD.scala:331)
  at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:92)
  at org.apache.spark.TaskContext.runTaskWithListeners(TaskContext.scala:161)
  at org.apache.spark.scheduler.Task.run(Task.scala:139)

Would appreciate any help. Thank you.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hi @JensonChoi. I will look at this today.

}
}

test("SparkToColumnar override node name for columnar input") {
withSQLConf(
SQLConf.ADAPTIVE_EXECUTION_ENABLED.key -> "true",
SQLConf.PARQUET_VECTORIZED_READER_ENABLED.key -> "true",
SQLConf.USE_V1_SOURCE_LIST.key -> "",
CometConf.COMET_NATIVE_SCAN_ENABLED.key -> "false",
CometConf.COMET_CONVERT_FROM_PARQUET_ENABLED.key -> "true",
CometConf.COMET_SHUFFLE_MODE.key -> "jvm") {
spark
.range(1000)
.selectExpr("id as key", "id % 8 as value")
.toDF("key", "value")
.write
.mode("overwrite")
.parquet("/tmp/test")
val df = spark.read.parquet("/tmp/test")
val rowDf = df.toDF()
rowDf.collect()

val planAfter = rowDf.queryExecution.executedPlan
assert(planAfter.toString.startsWith("AdaptiveSparkPlan isFinalPlan=true"))
val adaptivePlan = planAfter.asInstanceOf[AdaptiveSparkPlanExec].executedPlan
val nodeNames = adaptivePlan.collect { case c: CometSparkToColumnarExec =>
c.nodeName
}
assert(nodeNames.length == 1)
assert(nodeNames.head == "CometSparkColumnarToColumnar")
}
}

test("aggregate window function for all types") {
val numValues = 2048

Expand Down
Loading