tests: add benchmarks for log.values, log.has(), log.get()

[mistakia]

Two sets of benchmarks for log.values, log.has(), log.get().

Interesting results for the log.values benchmark with waves.

~60% drop in throughput from the initial wave (1000 entries -> 2000 entries). By wave no. 3 (4000 entries), it is down ~83% from the initial wave. By wave no. 14 (150000 entries), it starts to level off at ~97% drop from initial wave.

Machine: MacBook Pro (Mid 2012) / 2.6 GHz Intel Core i7 / 16 GB 1600 MHz DDR3

Related: #136

[haadcode]

Thanks for the PR @mistakia ❤️These are great additions to the benchmarking suite 👍

I was wondering what your idea was to separate the "single entry" test from "multiple entries" (*-plus.js) tests? Would they not work (benchmark) the same (ie. plus benchmark ran with arg 1)?

[mistakia]

I made the single entry test first (by quickly cloning the other benchmarks), then wanted another benchmark to see op/s at various log sizes.

Then I never realized that the initial wave of the *-plus.js benchmarks ran with arg 1 are identical to the single entry tests. ¯\_(ツ)_/¯

I think 10 seconds of the initial wave are sufficient to get rid of the single entry tests. If need be, I could add another argument to set the length in time of each wave.

[mistakia]

Probably best to close this in deference to #143

[haadcode]

Probably best to close this in deference to #143

Makes sense. Feel free to close it when you're ready.