All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=783699

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=feff49e12b268c549d540bf37b23a65432261866b0e476d43c3af5262a0e62b5


Bot(s) for this bug's original alert(s):

chromium-rel-win7-gpu-ati

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8963319407897438048

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : memory.long_running_idle_gmail_background_tbmv2
  Metric       : v8-gc-total_max/v8-gc-total_max

Revision             Result                  N
chromium@514493      67.7814 +- 76.0128      21      good
chromium@514593      69.7553 +- 74.9254      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests memory.long_running_idle_gmail_background_tbmv2

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8963319407897438048


For feedback, file a bug with component Speed>Bisection

The ref looks like pure bimodal noise.  No chance of bisecting that, perhaps we should remove this test?

+Ulan, should we consider disabling the ref build for this test/bot? Agreed w/Alex it is super noisy and just makes the baseline hard to see.

Ref build should become less noisy with the future updates.

I think ref is still useful for other metrics in this benchmark. Example:
https://chromeperf.appspot.com/report?sid=985e4f7cbb8af0cae117e05259167feeb1789cbf0a26fdd5902505076f36f5c8&start_rev=510632&end_rev=521230

alexclarke@, did this alert appear in general perf sheriff page? I thought that alerts on this metrics go to V8 memory sheriff rotation.

Improvements appear on https://chromeperf.appspot.com/alerts if you press show improvements.  Big ones are often bugs which is why I look out for them.

Thanks, Ulan! I think we can close this out.