Microbenchmarks for the go programming language
Benchmarks are found in their respective _test
file, e.g. select
benchmarks are in select_test.go
Benchmarks for reference, or stuff that didn't fit somewhere else:
BenchmarkEmpty
: Benchmarks the empty for loop used in all other benchmarks
Benchmarks for the select
statement:
BenchmarkSelectDefaultAfterRead[One-Nine]
: Selects the default case after trying to read from[One-Nine]
unbuffered channelsBenchmarkSelectDefaultBeforeRead[One-Three]
: Selects the default case after trying to read from[One-Three]
unbuffered channels, but the default case is put before the other cases in the sourceBenchmarkSelectDefaultAfterRead[One-Three]Buffered
: Selects the default case after trying to read from[One-Three]
buffered (1) channelsBenchmarkSelectDefaultAfterWrite[One-Three]
: Selects the default case after trying to write to[One-Three]
unbuffered channelsBenchmarkSelectDefaultBeforeWrite[One-Three]
: Selects the default case after trying to write to[One-Three]
unbuffered channels, but the default case is put before the other cases in the sourceBenchmarkSelectDefaultAfterWrite[One-Three]Buffered
: Selects the default case after trying to write to[One-Three]
buffered (1) channels
For results, see RESULTS.md
.
The results may change significantly (e.g. when I run them on a PC vs. when I run them on a laptop), but the interesting relations between results should remain the same.