Page MenuHomePhabricator

Focus on the users that have the worst performance
Closed, ResolvedPublic

Description

I think this could be a coming goal for us:

Focus on the users that have the the worst performance, setting up the goal to make Wikipedia faster for the 90 percentile.

Before we can set a measurable goal, we should focus on identifying the 90 percentile:

  • mobile/desktop?
  • connection type?
  • identify the largest problem for those users

Event Timeline

Peter created this task.Feb 1 2016, 11:31 AM
Peter updated the task description. (Show Details)
Peter raised the priority of this task from to Low.
Peter added a project: Performance-Team.
Peter added a subscriber: Peter.
Restricted Application added a subscriber: Aklapper. · View Herald TranscriptFeb 1 2016, 11:31 AM
ori renamed this task from Focus on the users that has the worst performance to Focus on the users that have the worst performance.Feb 1 2016, 6:54 PM
ori updated the task description. (Show Details)
ori set Security to None.
ori added a subscriber: ori.Feb 1 2016, 7:12 PM

One issue with this is that the world is not at a standstill. The 90th percentile is a moving target. The average internet connection speed in India went up by 23% in 2015:

The work we invest in characterizing and optimizing for the bottom 10% will be very quickly obsolete. For a small team such as ours, this is a serious matter.

Krinkle added a subscriber: Krinkle.Feb 1 2016, 7:54 PM

We should also remove our filter for the maximum of 60000 for Navigation Timing properties:

https://github.com/wikimedia/operations-puppet/blob/77d4eb8715/modules/webperf/files/navtiming.py#L114

This was added way way back. This shouldn't be needed with our percentiles and various other sanity filters (e.g. properties in order etc.)

Gilles added a subscriber: Gilles.EditedFeb 1 2016, 9:55 PM

That data source (stateoftheinternet) lacks distribution and a mobile/desktop breakdown. Typically we could be seeing a huge increase in bandwidth for desktop in metro areas because fiber is getting rolled out, while for rural areas mobile access isn't seeing any improvement.

To me the main questions are:

  • if we focused on things from this angle, how would the highest priority tasks be different than if we didn't?
  • how can energy spent on this not yield returns for higher bandwidth situations too?

I don't think we can really go wrong here, I would be surprised if there's anything we work on that makes a huge difference for the worst performing clients that wouldn't make performance better for everyone.

Krinkle closed this task as Resolved.
Krinkle claimed this task.

T125381 was resolved. We also have NavTiming breakdown by platform (desktop/mobile) and country. Next is to set up monitoring.