[Design] [D/f] pycon dogfood feedback

[Katie Capps Parlante]

Yes, we do need to be looking at RAM footprint.
http://wiki.osafoundation.org/bin/view/Projects/PerformanceProject#Memory_Footprint

A reminder, though, that we're focusing on performance starting in the 
next milestone (0.7alpha3). We have time allocated on people's schedules 
  then, and we'll have targets based on top-down 0.7 use cases. (We'll 
also have some targets based on stepping up to a larger repository, not 
necessarily based on 0.7 use cases).

If you are dogfooding Chandler, or if you use calendars and task lists, 
don't forget to respond to the survey, as this is going to be fodder for 
picking those targets:
http://lists.osafoundation.org/pipermail/design/2006-March/004308.html

For an overview (from a dev perspective) of how we're staging work 
across the alpha milestones, see the tables on the team pages:
http://wiki.osafoundation.org/bin/view/Projects/ApplicationProject
http://wiki.osafoundation.org/bin/view/Projects/DeveloperPlatformProject

Cheers,
Katie

Philippe Bossut wrote:
> Ted Leung wrote:
>> One other thing that I observed was that after I first setup to 
>> dogfood (create several calendars, import 2 calendars, share 1 
>> calendar), Chandler's virtual memory image was around 1 Gb of RAM.
>> Perhaps we also need to be thinking about RAM footprint targets.
> +1
> 
> I suggest we add memory footprint analysis as part of the perf 
> tools/data. Not sure we want to set memory targets (sounds like 
> something we'll never agree upon...) but detecting pathological use of 
> memory is surely a good thing.
> 
> Heikki, what do you think?
> 
> - Philippe
> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
> 
> Open Source Applications Foundation "Design" mailing list
> http://lists.osafoundation.org/mailman/listinfo/design