Page MenuHomePhabricator

Benchmark hot cache load times
Closed, DeclinedPublic

Description

Right now, all of our benchmarks are done with a completely cold cache. Changes like T90732 should intuitively reduce the load time on a cold cache, but it is not as obvious what it will do when the RL cache is hot. We should benchmark both scenarios so we can measure this.

Event Timeline

Catrope created this task.Feb 22 2015, 6:31 AM
Catrope assigned this task to ori.
Catrope raised the priority of this task from to Needs Triage.
Catrope updated the task description. (Show Details)
Catrope added a subscriber: Catrope.
Restricted Application added a subscriber: Aklapper. · View Herald TranscriptFeb 22 2015, 6:31 AM
Jdforrester-WMF triaged this task as Medium priority.Feb 23 2015, 6:16 PM
Jdforrester-WMF moved this task from To Triage to Bug Fixes on the VisualEditor board.
Jdforrester-WMF set Security to None.
Jdforrester-WMF removed ori as the assignee of this task.Sep 16 2015, 8:08 PM
Jdforrester-WMF added a subscriber: ori.
Jdforrester-WMF added a subscriber: Jdforrester-WMF.

This isn't actually specific to VE, so moving over to Performance to decide what they wish. Is it still relevant?

What is that supposed to be a link to? It's definitely not T90732.

Krinkle added a subscriber: Krinkle.EditedSep 16 2015, 8:16 PM

What benchmark?

jhsoby added a subscriber: jhsoby.Sep 18 2015, 10:17 AM

What is that supposed to be a link to? It's definitely not T90732.

I would assume that Catrope was referring to T90372, not T90732.

ori closed this task as Declined.Sep 21 2015, 6:04 PM
ori claimed this task.

Part of the VE performance sprint which ended.