perf: Add benchmarks for Spark Scan + Comet Exec (apache#863)

* Add benchmarks for Spark Scan + Comet Exec * address feedback * address feedback * revert removing env var from usage examples * fix
comphead · Aug 23, 2024 · cff7697 · cff7697
1 parent 3b497a2
commit cff7697
Show file tree

Hide file tree

Showing 2 changed files with 25 additions and 0 deletions.
diff --git a/spark/src/test/scala/org/apache/spark/sql/benchmark/CometTPCDSMicroBenchmark.scala b/spark/src/test/scala/org/apache/spark/sql/benchmark/CometTPCDSMicroBenchmark.scala
@@ -124,6 +124,20 @@ object CometTPCDSMicroBenchmark extends CometTPCQueryBenchmarkBase {
           cometSpark.sql(queryString).noop()
         }
       }
+      benchmark.addCase(s"$name$nameSuffix: Comet (Exec)") { _ =>
+        withSQLConf(
+          CometConf.COMET_ENABLED.key -> "true",
+          CometConf.COMET_NATIVE_SCAN_ENABLED.key -> "false",
+          CometConf.COMET_CONVERT_FROM_PARQUET_ENABLED.key -> "true",
+          CometConf.COMET_EXEC_SHUFFLE_ENABLED.key -> "true",
+          CometConf.COMET_SHUFFLE_MODE.key -> "auto",
+          CometConf.COMET_REGEXP_ALLOW_INCOMPATIBLE.key -> "true",
+          // enabling COMET_EXPLAIN_NATIVE_ENABLED may add overhead but is useful for debugging
+          CometConf.COMET_EXPLAIN_NATIVE_ENABLED.key -> "false",
+          CometConf.COMET_EXEC_ENABLED.key -> "true") {
+          cometSpark.sql(queryString).noop()
+        }
+      }
       benchmark.run()
     }
   }

diff --git a/spark/src/test/scala/org/apache/spark/sql/benchmark/CometTPCQueryBenchmarkBase.scala b/spark/src/test/scala/org/apache/spark/sql/benchmark/CometTPCQueryBenchmarkBase.scala
@@ -79,6 +79,17 @@ trait CometTPCQueryBenchmarkBase extends SqlBasedBenchmark with CometTPCQueryBas
           cometSpark.sql(queryString).noop()
         }
       }
+      benchmark.addCase(s"$name$nameSuffix: Comet (Exec)") { _ =>
+        withSQLConf(
+          CometConf.COMET_ENABLED.key -> "true",
+          CometConf.COMET_EXEC_ENABLED.key -> "true",
+          CometConf.COMET_NATIVE_SCAN_ENABLED.key -> "false",
+          CometConf.COMET_CONVERT_FROM_PARQUET_ENABLED.key -> "true",
+          CometConf.COMET_EXEC_SHUFFLE_ENABLED.key -> "true",
+          CometConf.COMET_SHUFFLE_MODE.key -> "auto") {
+          cometSpark.sql(queryString).noop()
+        }
+      }
       benchmark.run()
     }
   }