Confirmed users
431
edits
Performance: Difference between revisions
→Active
(→Active) |
|||
| Line 108: | Line 108: | ||
|work={{bug|815354}} | |work={{bug|815354}} | ||
|dependencies=OMTC | |dependencies=OMTC | ||
|notes= | |||
}} | |||
---- | |||
====<span style="color: green">Network Cache Rewrite</span>==== | |||
{{PerfProject | |||
|impact={{desktop}}{{mobile}}{{b2g}} | |||
|team=Honza Bambas, Michal Novotny, Jason Duell (manager) | |||
|duration=TBD | |||
|goal=A cache that causes no main thread I/O jank (directly or indirectly via locks)<sup>1</sup>, and that avoids losing the entire cache during crashes or unclean shutdown.<sup>2</sup><br> | |||
<sup>[1]</sup><span style="font-size: 80%;">Currently the cache causes jank (sometimes in excess of 1 second) via main thread waiting while cache thread holds lock and does I/O.</span><br> | |||
<sup>[2]</sup><span style="font-size: 80%;">Cache is currently thrown away unless clean shutdown. Cache loss rates per startup are 9% on windows desktop, 11% OSX, 22% Linux desktop, 25% metro. New design will keep all cache entries and detect corruption on a per-entry basis.</span><br/> | |||
|reports= | |||
|work={{bug|877301}} | |||
|dependencies= | |||
|notes= | |notes= | ||
}} | }} | ||