All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=844449

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=bf06af74b5f1a4dfcee3fccb5032b9ea1270a0f8eb6cce6dcb5d5bd9bcdef7d1


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

android-webview-nexus6

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/12af318c240000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/12af318c240000

NetworkService: Support subresource requests to extension service workers. by falken@chromium.org
https://chromium.googlesource.com/chromium/src/+/62798c60492048987a8c95b986a5f56765fb448b

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

I don't think this is my commit. That only affects service workers and the tests in question (blob-perf) don't seem to use service workers (furthermore, it only affects extension service workers when NetworkService is enabled).

It looks like the graph merely returned to May 1st levels. Somehow the test got fast around May 1st but I don't think there was an intentional change at that time. I suspect it's just noise. Leaving as Untriaged for OWP Storage team.

Yeah, agreed that this is probably just noise.