Benchmarking flat_sets.
Reported by
dyaros...@yandex-team.ru,
Jan 18 2017
|
||||||
Issue descriptionThis task is a followup to https://bugs.chromium.org/p/chromium/issues/detail?id=671759 While developing flat_sets, we based on benchmarks for HistoryQuickProvider added here https://bugs.chromium.org/p/chromium/issues/detail?id=643668 We have to develop benchmarks for different scenarios using sets and compare performance of flat_set/std::set for these scenarios in order to come up with guidance for where to use flat_sets vs std::sets. Another thing to benchmark ought to be gather everything in a vector and sort + unique where it makes sense. This is a useful baseline to measure inserting many elements.
,
Jan 19 2017
ajha: why did you add those labels? This is not a general metrics task. This will not involve changing any metrics internals, and may live somewhere else (say, telemetry). Also, I don't see why this needs triage help; this is a follow-up from a changelist in progress that the filer will do when the changelist is submitted.
,
Jan 19 2017
Thanks mark@ for update. Adjusting to more appropriate bug component as per C#2.
,
Jan 26 2017
,
Jan 26 2017
,
Feb 12 2018
This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue. Sorry for the inconvenience if the bug really should have been left as Available. If you change it back, also remove the "Hotlist-Recharge-Cold" label. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Feb 13 2018
Guidance was written by brettw
,
Feb 13 2018
@danakj Can you give me a link please? Benchmarking such things is a huge pain, would be terribly nice to have a library btw. But reviewers were against it at the time. I actually did some benchmarking on the side - everything except for insert(first, last) is ok. There are 3 calls to insert(first, last) in the entire code base, so it doesn't seem to matter. And implementing it properly is a bigger pain than Chromium needs, in my opinion.
,
Feb 13 2018
The guidance is in base/containers/README.md If you'd like to do more then you can re-open this and we can assign it to you. It's been idle for a year tho.
,
Feb 13 2018
@danakj - Oh I see - usage advice, not benchmarking advice. No, I don't see much value in it now. I just misunderstood, that you wanted to reopen this ticket. My bad. |
||||||
►
Sign in to add a comment |
||||||
Comment 1 by ajha@chromium.org
, Jan 19 2017Labels: TE-NeedsTriageHelp