siknezevic commented on a change in pull request #27246:
URL: https://github.com/apache/spark/pull/27246#discussion_r445307223
##########
File path:
sql/core/benchmarks/ExternalAppendOnlyUnsafeRowArrayBenchmark-results.txt
##########
@@ -42,4 +42,8 @@ Spilling with 10000 rows: Best Time(ms) Avg
Time(ms) Stdev(m
UnsafeExternalSorter 11 11
1 14.7 68.0 1.0X
ExternalAppendOnlyUnsafeRowArray 9 10
1 17.1 58.5 1.2X
-
+OpenJDK 64-Bit Server VM 1.8.0_252-8u252-b09-1~16.04-b09 on Linux
4.4.0-178-generic
+Intel(R) Xeon(R) CPU E5-2687W v3 @ 3.10GHz
+Spilling SpillReader with 16000 rows: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
+------------------------------------------------------------------------------------------------------------------------
+UnsafeSorterSpillReader_bufferSize1024 411 426
13 0.6 1607.2 1.0X
Review comment:
There are three spill files. It will result in three I/Os for each
created iterator, because it has to read number of rows from each spilled file.
So, we see more than 2000 X performance difference.
----------------------------------------------------------------
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:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
