1
100%
master: 100%

Ran 27 May 2016 08:25AM UTC

Files 65

Run time 3s

Badge

Embed ▾

Committed 27 May 2016 06:56AM UTC coverage: 94.482% (-0.02%) from 94.5%

Job # TERM=dumb

Build Type

push

travis-ci

Committed by

ben-manes

Commit Message

Adjust adaptive W-TinyLFU by larger amounts

Adapting by +/- 1 helped recency skewed workloads, but never grew the
window large enough to be optimal. By using a larger increment the
window grows quicker to a closer optimum. The policy will still adapt
towards frequency, but using a smaller decrement helps bias it longer
towards recency. As this only matters for small traces, it seems to
be a good trade-off that results in better hit races in recency skewed
without hurting frequency skewed (~0.5% at worst).

Coverage Stats

5325 of 5636 relevant lines covered (94.48%)

0.94 hits per line

ben-manes / caffeine / 1767 / 1
100%
master: 100%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 1767.1 (TERM=dumb)

ben-manes / caffeine / 1767 / 1 100% master: 100%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 1767.1 (TERM=dumb)

ben-manes / caffeine / 1767 / 1
100%
master: 100%

README BADGES
x