[SPARK-29320][TESTS] Compare sql/core module in JDK8/11 (Part 1)

[dongjoon-hyun]

What changes were proposed in this pull request?

This PR regenerates the following sql/core benchmarks in JDK8/11 to compare the result.
(These are less time-consuming half of the all sql/core benchmarks)

AggregateBenchmark
BloomFilterBenchmark
BuiltInDataSourceWriteBenchmark
CSVBenchmark
ColumnarBatchBenchmark
CompressionSchemeBenchmark
DataSourceReadBenchmark
DatasetBenchmark
DateTimeBenchmark
ExternalAppendOnlyUnsafeRowArrayBenchmark
ExtractBenchmark
FilterPushdownBenchmark
HashedRelationMetricsBenchmark
InExpressionBenchmark
JoinBenchmark
MakeDateTimeBenchmark
MiscBenchmark
OrcNestedSchemaPruningBenchmark
OrcV2NestedSchemaPruningBenchmark
ParquetNestedSchemaPruningBenchmark
PrimitiveArrayBenchmark
RangeBenchmark
SortBenchmark
UDFBenchmark
UnsafeArrayDataBenchmark
WideSchemaBenchmark
WideTableBenchmark

Why are the changes needed?

According to the result, there are some difference between JDK8/JDK11. For example, ByteBuffer API and DirectByteBuffer performance is switched like the following.

JDK8

Int Read/Write:                           Best Time(ms)   Avg Time(ms)   Stdev(ms)    Rate(M/s)   Per Row(ns)   Relative
 ------------------------------------------------------------------------------------------------------------------------
Java Array                                          245            245           0       1338.6           0.7       1.0X
ByteBuffer Unsafe                                   445            445           0        736.4           1.4       0.6X
ByteBuffer API                                     1936           1938           3        169.2           5.9       0.1X
DirectByteBuffer                                    743            744           1        440.8           2.3       0.3X

JDK11

Int Read/Write:                           Best Time(ms)   Avg Time(ms)   Stdev(ms)    Rate(M/s)   Per Row(ns)   Relative
 ------------------------------------------------------------------------------------------------------------------------
Java Array                                          232            234           1       1409.7           0.7       1.0X
ByteBuffer Unsafe                                   396            396           0        827.9           1.2       0.6X
ByteBuffer API                                      602            603           0        544.0           1.8       0.4X
DirectByteBuffer                                   1185           1186           2        276.6           3.6       0.2X

Does this PR introduce any user-facing change?

No.

How was this patch tested?

This is a test-only PR. We need to run benchmark.

[SparkQA]

Test build #111650 has finished for PR 25992 at commit 98bb243.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[HyukjinKwon]

Woah .. huge PR but +1 from a cursory look

[maropu]

Where these result did go anywhere? We need to keep this empty file?

[dongjoon-hyun]

Oh, this seems to be truncated mistakenly for some reason.

[maropu]

oh, yea.

[dongjoon-hyun]

After resolve the conflicts, I'll regenerate the result and reopen this.

[dongjoon-hyun]

Oops. Rebasing and force-pushing seems to lose the track.

[dongjoon-hyun]

#26003 is made.