Github user cloud-fan commented on a diff in the pull request:
https://github.com/apache/spark/pull/16391#discussion_r93788919
--- Diff:
sql/core/src/test/scala/org/apache/spark/sql/DatasetBenchmark.scala ---
@@ -170,36 +176,39 @@ object DatasetBenchmark {
val benchmark3 = aggregate(spark, numRows)
/*
- OpenJDK 64-Bit Server VM 1.8.0_91-b14 on Linux
3.10.0-327.18.2.el7.x86_64
- Intel Xeon E3-12xx v2 (Ivy Bridge)
+ Java HotSpot(TM) 64-Bit Server VM 1.8.0_60-b27 on Mac OS X 10.12.1
+ Intel(R) Core(TM) i7-4960HQ CPU @ 2.60GHz
+
back-to-back map: Best/Avg Time(ms)
Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
- RDD 3448 / 3646 29.0
34.5 1.0X
- DataFrame 2647 / 3116 37.8
26.5 1.3X
- Dataset 4781 / 5155 20.9
47.8 0.7X
+ RDD 3963 / 3976 25.2
39.6 1.0X
+ DataFrame 826 / 834 121.1
8.3 4.8X
+ Dataset 5178 / 5198 19.3
51.8 0.8X
--- End diff --
for "back-to-back map", the logic is so simple that the code generated by
`Dataset` is less efficient than `RDD`. `RDD` just adds 1 to the input `Long`,
the only overhead is boxing, while `Dataset` generates code like this:
```
boolean mapelements_isNull = true;
long mapelements_value = -1L;
if (!false) {
mapelements_argValue = range_value;
mapelements_isNull = false;
if (!mapelements_isNull) {
Object mapelements_funcResult = null;
mapelements_funcResult = mapelements_obj.apply(mapelements_argValue);
if (mapelements_funcResult == null) {
mapelements_isNull = true;
} else {
mapelements_value = (Long) mapelements_funcResult;
}
}
}
```
`Dataset` still has the boxing overhead, but its code is more verbose. And
`Dataset` has to write the long to un unsafe row at last, which is another
overhead. These are the reasons why `Dataset` is slower than `RDD` for this
simple case.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
