Avoid alloc on creating hashable func

[carsonip]

benchstat:

goos: linux
goarch: amd64
pkg: github.com/elastic/apm-aggregation/aggregators
cpu: 11th Gen Intel(R) Core(TM) i7-11800H @ 2.30GHz
                            │ bench-out/hashable-func-alloc-before │ bench-out/hashable-func-alloc-after │
                            │                sec/op                │   sec/op     vs base                │
AggregateCombinedMetrics-16                            4.003µ ± 3%   4.072µ ± 2%        ~ (p=0.342 n=10)
AggregateBatchSerial-16                                10.46µ ± 3%   10.26µ ± 4%        ~ (p=0.645 n=10)
AggregateBatchParallel-16                              11.07µ ± 5%   10.73µ ± 5%        ~ (p=0.190 n=10)
CombinedMetricsEncoding-16                             4.842µ ± 2%   4.829µ ± 3%        ~ (p=0.869 n=10)
CombinedMetricsDecoding-16                             4.449µ ± 2%   4.494µ ± 2%        ~ (p=0.247 n=10)
CombinedMetricsToBatch-16                              417.2µ ± 3%   418.9µ ± 2%        ~ (p=0.912 n=10)
EventToCombinedMetrics-16                             1007.0n ± 2%   845.4n ± 2%  -16.05% (p=0.000 n=10)
geomean                                                8.832µ        8.588µ        -2.77%

                            │ bench-out/hashable-func-alloc-before │  bench-out/hashable-func-alloc-after   │
                            │                 B/op                 │     B/op      vs base                  │
AggregateCombinedMetrics-16                         4.862Ki ± 2%     4.868Ki ± 1%        ~ (p=0.956 n=10)
AggregateBatchSerial-16                             11.48Ki ± 1%     11.34Ki ± 0%   -1.17% (p=0.000 n=10)
AggregateBatchParallel-16                           11.51Ki ± 1%     11.32Ki ± 1%   -1.67% (p=0.000 n=10)
CombinedMetricsEncoding-16                            0.000 ± 0%       0.000 ± 0%        ~ (p=1.000 n=10) ¹
CombinedMetricsDecoding-16                          10.01Ki ± 0%     10.01Ki ± 0%        ~ (p=0.593 n=10)
CombinedMetricsToBatch-16                           37.66Ki ± 0%     37.66Ki ± 0%        ~ (p=1.000 n=10) ¹
EventToCombinedMetrics-16                             80.00 ± 0%       16.00 ± 0%  -80.00% (p=0.000 n=10)
geomean                                                          ²                 -20.85%                ²
¹ all samples are equal
² summaries must be >0 to compute geomean

                            │ bench-out/hashable-func-alloc-before │ bench-out/hashable-func-alloc-after  │
                            │              allocs/op               │ allocs/op   vs base                  │
AggregateCombinedMetrics-16                           39.00 ± 3%     39.00 ± 3%        ~ (p=1.000 n=10)
AggregateBatchSerial-16                               79.00 ± 0%     67.00 ± 1%  -15.19% (p=0.000 n=10)
AggregateBatchParallel-16                             79.00 ± 0%     67.00 ± 0%  -15.19% (p=0.000 n=10)
CombinedMetricsEncoding-16                            0.000 ± 0%     0.000 ± 0%        ~ (p=1.000 n=10) ¹
CombinedMetricsDecoding-16                            40.00 ± 0%     40.00 ± 0%        ~ (p=1.000 n=10) ¹
CombinedMetricsToBatch-16                             308.0 ± 0%     308.0 ± 0%        ~ (p=1.000 n=10) ¹
EventToCombinedMetrics-16                             6.000 ± 0%     2.000 ± 0%  -66.67% (p=0.000 n=10)
geomean                                                          ²               -18.46%                ²
¹ all samples are equal
² summaries must be >0 to compute geomean

[lahsivjar]

Nice 🎉 ! I was thinking of writing a protoc generator to generate these hash functions, my brain didn't even think that we can just add the hash functions in the generated proto directory 🤦

[lahsivjar]

I think we can change nullable.NullableBool to nullable.Bool