~/talk

How to Performance

Talkat Full Stack Fest 2015 · 2015-09-01..09-05· [ruby, web-performance]
slides:none on file

A reproducible test app and a Benchmark IPS harness establish a baseline, then profiling with RubyProf and sampling with StackProf (switching CPU→wall time) plus Ruby VM stats and AllocationTracer finds real culprits: an unnecessary URL_helpers include that busts the method cache, hot delegate code, and heavy string allocations. Iterations move from ~393 to ~828, integration tests go from ~2.53× slower to about 1.2× slower versus controller tests, and targeted fixes (move include, memoize, freeze strings) produce most of the gains.

AI-written summary · about

  1. 0:00Introduction
  2. 2:22Benchmarking vs Profiling
  3. 4:43Building a Baseline
  4. 9:52Benchmarking
  5. 11:08Profilers
  6. 14:32Ruby prof
  7. 15:21Stock prof
  8. 18:41Walltime
  9. 20:21Method Cash
  10. 22:45Method Cash 2
  11. 27:07String Allocation
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.