size in log(n)

[matthewleon]

addresses #100
adds benchmarks, addressing #101

Before:

size: singleton map
mean   = 7.85 μs
stddev = 62.54 μs
min    = 1.91 μs
max    = 1.60 ms
size: small map (100)
mean   = 303.09 μs
stddev = 482.41 μs
min    = 205.12 μs
max    = 3.85 ms
size: midsize map (10000)
mean   = 40.37 ms
stddev = 1.42 ms
min    = 36.60 ms
max    = 46.82 ms
size: big map (1000000)
mean   = 5.79 s
stddev = 60.76 ms
min    = 5.67 s
max    = 5.89 s

After:

size: singleton map
mean   = 3.00 μs
stddev = 38.96 μs
min    = 326.00 ns
max    = 1.16 ms
size: small map (100)
mean   = 10.16 μs
stddev = 7.21 μs
min    = 8.35 μs
max    = 175.09 μs
size: midsize map (10000)
mean   = 901.31 μs
stddev = 198.26 μs
min    = 822.48 μs
max    = 1.88 ms
size: big map (1000000)
mean   = 96.70 ms
stddev = 2.00 ms
min    = 92.99 ms
max    = 99.69 ms

[paf31]

Looks great, and thanks for adding the benchmarks!

@garyb What do you think about moving minibench into core?

[garyb]

What do you think about moving minibench into core?

💯 👍

[paf31]

Cool, I'll take care of that later today.

[garyb]

Nice, looking forward to having benchmarking on all the important stuff 😄

[paf31]

I just realized this is actually O(n), which I documented in the release notes 😄

Thanks again!

[matthewleon]

Good point!

…

On Mon, 5 Jun 2017 at 21:59, Phil Freeman ***@***.***> wrote: I just realized this is actually O(n), which I documented in the release notes 😄 Thanks again! — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#107 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AARBSynTOX-tpvpQuaJHbS-Vy_pEEH0Tks5sBGwrgaJpZM4Nv61l> .