gortiz commented on PR #12354:
URL: https://github.com/apache/pinot/pull/12354#issuecomment-1948104199
I've applied similar changes in SUM and results are:
In AMD Ryzen 9 3900X, Java 11:
```
Benchmark (_aQueryTemplate)
(_impl) (_nullHandling) Mode Cnt Score Error Units
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
normal true thrpt 25 1.868 ± 0.153 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
normal false thrpt 25 1.868 ± 0.172 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldDouble true thrpt 25 1.884 ± 0.145 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldDouble false thrpt 25 1.841 ± 0.175 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldPrimitive true thrpt 25 1.942 ± 0.153 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldPrimitive false thrpt 25 1.869 ± 0.159 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldHolder true thrpt 25 1.918 ± 0.165 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldHolder false thrpt 25 1.941 ± 0.150 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
normal true thrpt 25 1.911 ± 0.178 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
normal false thrpt 25 1.879 ± 0.162 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldDouble true thrpt 25 1.869 ± 0.171 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldDouble false thrpt 25 1.856 ± 0.157 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldPrimitive true thrpt 25 1.884 ± 0.196 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldPrimitive false thrpt 25 1.915 ± 0.172 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldHolder true thrpt 25 1.922 ± 0.167 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldHolder false thrpt 25 1.966 ± 0.172 ops/ms
```
In AMD Ryzen 9 3900X, Java 21:
```
Benchmark (_aQueryTemplate)
(_impl) (_nullHandling) Mode Cnt Score Error Units
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
normal true thrpt 25 2.477 ± 0.091 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
normal false thrpt 25 2.509 ± 0.074 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldDouble true thrpt 25 2.609 ± 0.064 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldDouble false thrpt 25 2.501 ± 0.083 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldPrimitive true thrpt 25 2.652 ± 0.106 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldPrimitive false thrpt 25 2.627 ± 0.095 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldHolder true thrpt 25 2.664 ± 0.087 ops/ms
BenchmarkSumAggregation.test select sum(valueLong, '%s') from benchmark
foldHolder false thrpt 25 2.701 ± 0.097 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
normal true thrpt 25 2.428 ± 0.072 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
normal false thrpt 25 2.450 ± 0.049 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldDouble true thrpt 25 2.451 ± 0.073 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldDouble false thrpt 25 2.503 ± 0.055 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldPrimitive true thrpt 25 2.573 ± 0.067 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldPrimitive false thrpt 25 2.567 ± 0.084 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldHolder true thrpt 25 2.664 ± 0.081 ops/ms
BenchmarkSumAggregation.test select sum(valueInt, '%s') from benchmark
foldHolder false thrpt 25 2.622 ± 0.074 ops/ms
```
--
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.
To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org
For additional commands, e-mail: commits-h...@pinot.apache.org
