I think the graph are:
https://chromeperf.appspot.com/report?sid=d01fe203246bb57c8b8f6c0b5ca5fa64898d7099f1f61e2e149d20d32738ac08

right, that's

ChromiumChrome/Google Chrome Mac/sizes


is that better than 

ChromiumPerf/chromium_perf/sizes


?  (eyaich@ halp! :)

(I think ChromiumChrome/Google Chrome Mac/ has continuous data, but ChromiumPerf/chromium_perf/ does not. My reading of the goals in Issue 758326 is that perf stuff is getting consolidated to chromium_perf rather than having a bunch of platform-specific ~folders like "/Google Chrome Mac/sizes/" and "/mac/sizes")

So afaik this isn't related to any changes I made under go/sdvzd (Issue 758326).  We updated builder names here, but we didn't consolidate reporting to the waterfall level (ie I can't explain the switch from ChromiumPerf/mac/sizes to ChromiumPerf/chromium_perf/sizes)

I am slightly confused what the relationship is to ChromiumChrome/Google Chrome Mac/sizes.  That is a different waterfall than ChromiumPerf, so I am not sure how the graph you linked Ned could be the graph he is looking for.  We are still running the sizes on our mac builder on the chromium perf waterfall:

https://ci.chromium.org/p/chrome/builders/luci.chrome.ci/mac-builder-perf/11715

Ned or John would be the likely people to know if anything else has changed in this space with regards to Luci. 

Simon will be the one to make the changes on the monitoring side since that is the dashboard world.  

I don't have time to dig into why we have the switch either, but seems like the action item here is to enabling monitoring on ChromiumChrome/Google Chrome Mac/sizes/*

Reassign to Sean to triage since this is a perf dashboard request

switching to ChromiumChrome/Google Chrome Mac/sizes/* sgtm

but note that all platforms may be affected by this change [whatever it is]. We may have missed regressions there too.

i.e. All of ChromiumPerf/{linux,mac,win,win32}/sizes do not seem to be getting new data. They seem to have data in corresponding /Google Chrome Foo/ though.

https://chromeperf.appspot.com/report?sid=902e56b53ea6cd0845b2f42ec3e3ff359737810765cd8d0e354777b91cf56b9b&start_rev=578255&end_rev=598562

Updating subject to not mention ChromiumPerf/chromium_perf/sizes/ -- although this has all platforms combined, the data does not look good, e.g., for chrome.dll (link above).

has the dashboard been updated?

We have another, huge (10MB), regression from 10 days ago now. in Chromium.app this time.

Deleted: Screenshot from 2018-10-22 15-49-42.png 132 KB

	Screenshot from 2018-10-22 15-49-42.png 132 KB View Download

My guess for the 10MB regression is "Ship SwiftShader on MacOS" -> https://chromium-review.googlesource.com/c/chromium/src/+/1276550

Filed Issue 897573 for the new regression.

I added ChromiumChrome/*/sizes/* to the Chromium Perf Sheriff rotation.
If you want to unmonitor some of the measurements or bots, or migrate some old timeseries to new names, or anything else with the chromeperf dashboard, please file a separate bug directly to me. 

Assigning back to tapted to see if anything else needs to be done here.

Thanks Ben! Can think of anything else that needs to be done.