huaxingao commented on a change in pull request #34594:
URL: https://github.com/apache/spark/pull/34594#discussion_r748936264
##########
File path:
sql/core/src/test/scala/org/apache/spark/sql/execution/benchmark/BloomFilterBenchmark.scala
##########
@@ -81,8 +80,55 @@ object BloomFilterBenchmark extends SqlBasedBenchmark {
}
}
+ private def writeParquetBenchmark(): Unit = {
+ withTempPath { dir =>
+ val path = dir.getCanonicalPath
+
+ runBenchmark(s"Parquet Write") {
+ val benchmark = new Benchmark(s"Write ${scaleFactor}M rows", N, output
= output)
+ benchmark.addCase("Without bloom filter") { _ =>
+ df.write.mode("overwrite").parquet(path + "/withoutBF")
+ }
+ benchmark.addCase("With bloom filter") { _ =>
+ df.write.mode("overwrite")
+ .option(ParquetOutputFormat.BLOOM_FILTER_ENABLED + "#value", true)
+ .parquet(path + "/withBF")
+ }
+ benchmark.run()
+ }
+ }
+ }
+
+ private def readParquetBenchmark(): Unit = {
+ val blockSizes = Seq(2 * 1024 * 1024, 4 * 1024 * 1024, 6 * 1024 * 1024, 8
* 1024 * 1024,
+ 12 * 1024 * 1024, 16 * 1024 * 1024, 32 * 1024 * 1024)
+ for (blocksize <- blockSizes) {
+ withTempPath { dir =>
+ val path = dir.getCanonicalPath
+ df.write.option("parquet.block.size", blocksize).parquet(path +
"/withoutBF")
+ df.write.option(ParquetOutputFormat.BLOOM_FILTER_ENABLED + "#value",
true)
+ .option("parquet.block.size", blocksize)
+ .parquet(path + "/withBF")
+
+ runBenchmark(s"Parquet Read") {
+ val benchmark = new Benchmark(s"Read a row from ${scaleFactor}M
rows", N, output = output)
+ benchmark.addCase("Without bloom filter, blocksize: " + blocksize) {
_ =>
+ spark.read.parquet(path + "/withoutBF").where("value = 0").noop()
+ }
+ benchmark.addCase("With bloom filter, blocksize: " + blocksize) { _
=>
+ spark.read.option(ParquetInputFormat.BLOOM_FILTERING_ENABLED, true)
+ .parquet(path + "/withBF").where("value = 0").noop()
+ }
+ benchmark.run()
+ }
+ }
+ }
+ }
+
override def runBenchmarkSuite(mainArgs: Array[String]): Unit = {
- writeBenchmark()
- readBenchmark()
+ writeORCBenchmark()
+ readORCBenchmark()
+ writeParquetBenchmark()
+ readParquetBenchmark()
Review comment:
Will update the results file once Git Actions benchmark tests finish.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
