Results:

Results after just adding the benchmark test on commit: 5f8bf8a3fec64e2df6d1032381a6084028fd7867 (Base line no changes)

BenchmarkClientServerRoundTripSmallMsg-12 11257 105941 ns/op 17392 B/op 290 allocs/op BenchmarkClientServerRoundTripBigStrMsg-12 9408 127801 ns/op 71079 B/op 293 allocs/op BenchmarkClientServerRoundTripBigMapBigStrMsg-12 3555 309087 ns/op 187275 B/op 2725 allocs/op

BenchmarkClientServerRoundTripSmallMsg-12 10671 107704 ns/op 17393 B/op 290 allocs/op BenchmarkClientServerRoundTripBigStrMsg-12 8620 132313 ns/op 71168 B/op 293 allocs/op BenchmarkClientServerRoundTripBigMapBigStrMsg-12 3576 315491 ns/op 187284 B/op 2725 allocs/op

BenchmarkClientServerRoundTripSmallMsg-12 11174 110227 ns/op 17395 B/op 290 allocs/op BenchmarkClientServerRoundTripBigStrMsg-12 8658 139203 ns/op 71103 B/op 293 allocs/op BenchmarkClientServerRoundTripBigMapBigStrMsg-12 3556 315230 ns/op 187242 B/op 2725 allocs/op

Results after just adding the benchmark test on commit: dad755d6cb11f4daab738eec8133554a7c3c48e7 (pre-compiled names regex)

Just focusing on optimizing Allocations per Operation for this set of changes

BenchmarkClientServerRoundTripSmallMsg-12 11971 94264 ns/op 11850 B/op 208 allocs/op BenchmarkClientServerRoundTripBigStrMsg-12 9106 114902 ns/op 65520 B/op 211 allocs/op BenchmarkClientServerRoundTripBigMapBigStrMsg-12 3831 298895 ns/op 181501 B/op 2642 allocs/op

BenchmarkClientServerRoundTripSmallMsg-12 12870 92300 ns/op 11848 B/op 208 allocs/op BenchmarkClientServerRoundTripBigStrMsg-12 9277 114179 ns/op 65505 B/op 211 allocs/op BenchmarkClientServerRoundTripBigMapBigStrMsg-12 3537 293994 ns/op 181645 B/op 2642 allocs/op

BenchmarkClientServerRoundTripSmallMsg-12 12685 92157 ns/op 11848 B/op 208 allocs/op BenchmarkClientServerRoundTripBigStrMsg-12 9462 115722 ns/op 65519 B/op 211 allocs/op BenchmarkClientServerRoundTripBigMapBigStrMsg-12 3644 295176 ns/op 181563 B/op 2642 allocs/op