[SPARK-12888][SQL] benchmark the new hash expression

[cloud-fan]

Benchmark it on 4 different schemas, the result:

Intel(R) Core(TM) i7-4960HQ CPU @ 2.60GHz
Hash For simple:                   Avg Time(ms)    Avg Rate(M/s)  Relative Rate
-------------------------------------------------------------------------------
interpreted version                       31.47           266.54         1.00 X
codegen version                           64.52           130.01         0.49 X

Intel(R) Core(TM) i7-4960HQ CPU @ 2.60GHz
Hash For normal:                   Avg Time(ms)    Avg Rate(M/s)  Relative Rate
-------------------------------------------------------------------------------
interpreted version                     4068.11             0.26         1.00 X
codegen version                         1175.92             0.89         3.46 X

Intel(R) Core(TM) i7-4960HQ CPU @ 2.60GHz
Hash For array:                    Avg Time(ms)    Avg Rate(M/s)  Relative Rate
-------------------------------------------------------------------------------
interpreted version                     9276.70             0.06         1.00 X
codegen version                        14762.23             0.04         0.63 X

Intel(R) Core(TM) i7-4960HQ CPU @ 2.60GHz
Hash For map:                      Avg Time(ms)    Avg Rate(M/s)  Relative Rate
-------------------------------------------------------------------------------
interpreted version                    58869.79             0.01         1.00 X
codegen version                         9285.36             0.06         6.34 X

[SparkQA]

Test build #49621 has finished for PR 10816 at commit d753ce8.

• This patch fails RAT tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • class ALS(@Since(\"1.4.0\") override val uid: String) extends Estimator[ALSModel] with ALSParams

[cloud-fan]

The hash exprssion is slow on simple schema because the UnsafeProjection has some unnecessary opeations every round and will hurt the performance if each round is quite small(like simple schema), this can be fixed by #10809. For array type, still investigate.

cc @nongli

[cloud-fan]

retest this please

[nongli]

I'm confused by the conclusino. We don't use UnsafeProjection anymore right?

[cloud-fan]

We need UnsafeProjection to execute the codegen version of hash expression.

[SparkQA]

Test build #49629 has finished for PR 10816 at commit 1771733.

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

[nongli]

Can you include the results as comments for each test() case?

[nongli]

LGTM

[cloud-fan]

we should run this benchmark again after #10809 is merged.

[rxin]

I'm going to merge this. Please submit an update once #10809 is merged and put the result in the test case as comments.