compares vanilla racer to forks using docopt and clap

racer Speed Test.md

About

Some very naive tests comparing forks and vanilla phildawes/racer

• Uses current stable Rust (rustc 1.3.0 (9a92aaf19 2015-09-15))
  • All version compiled with --release
• Uses valgrind
  • valgrind racer complete std::io::B
• Uses Python 3.4.3
  • python -m timeit -n 1 -r 100 -s 'import os' 'os.system("racer complete std::io::B")'
• Uses time
  • for i in {0..10}; do time racer complete std::io::B; done | grep -v MATCH
  • Manual averaging
• All tests performed on Intel(R) Core(TM) i7-4710HQ CPU @ 2.50GHz
• OS: Arch Linux x64 (Kernel 4.2.2-1)

Vanilla

Uses branch phildawes/master#b9750c373a

• Number of Allocs: 12
• Bytes Allocated: 66,753
• Python: best of 100: 5.62 msec per loop
• Average time: r 0.00, u 0.00, s 0.004

Uses branch phildawes/master#1e04569993

• Number of Allocs: 1,198
• Bytes Allocated: 406,289
• Python: best of 100: 5.48 msec per loop
• Average time: r 0.00, u 0.00, s 0.007

Uses branch phildawes/master#22b452de

• Number of Allocs: 11,793
• Bytes Allocated: 2,582,216
• Python: best of 100: 9.75 msec per loop
• Average time: r 0.01, u 0.00, s 0.008

Docopt v0.6.70

Uses branch kbknapp/docopt#7bcb2ae2

• Number of Allocs: 32,186
• Bytes Allocated: 7,093,768
• Python: best of 100: 14 msec per loop
• Average time: r 0.01, u 0.00, s 0.013

clap

v1.4.5 Uses branch kbknapp/clap-update#6bd1bd8ca8

• Number of Allocs: 13
• Bytes Allocated: 67,185
• Python: best of 100: 4.71 msec per loop
• Average time: r 0.01, u 0.00, s 0.04

v1.3.1 Uses branch kbknapp/clap-update#68f5409b5d

• Number of Allocs: 1,358
• Bytes Allocated: 588,697
• Python: best of 100: 5.64 msec per loop
• Average time: r 0.01, u 0.00, s 0.07

v1.2.0 Uses branch kbknapp/clap#0450278b

• Number of Allocs: 11,969
• Bytes Allocated: 3,161,312
• Python: best of 100: 9.54 msec per loop
• Average time: r 0.01, u 0.00, s 0.08