Disabling Analysis of Table stats to comply with Benchmarking Spec

PiperOrigin-RevId: 564100634
GoogleCloudPlatform · Sep 10, 2023 · 5ef57f4 · 5ef57f4
1 parent 3c27041
commit 5ef57f4
Showing 1 changed file with 5 additions and 4 deletions.
diff --git a/perfkitbenchmarker/scripts/spark_sql_test_scripts/spark_table.py b/perfkitbenchmarker/scripts/spark_sql_test_scripts/spark_table.py
@@ -48,10 +48,11 @@ def main(args):
     # the Hive Metastore. I do not believe this interoperates with Hive's own
     # statistics. See
     # https://jaceklaskowski.gitbooks.io/mastering-spark-sql/content/spark-sql-LogicalPlan-AnalyzeColumnCommand.html
-    columns = ','.join(spark.table(table).columns)
-    spark.sql(
-        'ANALYZE TABLE {} COMPUTE STATISTICS FOR COLUMNS {}'.format(
-            table, columns))
+    # Disabling to comply with the TPC spec.
+    # columns = ','.join(spark.table(table).columns)
+    # spark.sql(
+    #     'ANALYZE TABLE {} COMPUTE STATISTICS FOR COLUMNS {}'.format(
+    #         table, columns))
 
 
 if __name__ == '__main__':