brainstorming for UDS-N - Performance

[Martin Pitt]

Wouter Stomp [2010-09-29  9:42 +0200]:
> Perhaps the 10 second boot time target that was originally set for
> Lucid could be pursued further? (the end-result in lucid was 17s)

I agree. That's something that would also interest me personally.

Indeed it seems that maverick dramatically regressed here, at least on
my end. In August I got a nice 8 second boot [1], which already wasn't
optimal (ureadahead shouldn't have blocked the boot), I
now have a 30 second boot [2], and that on a quad-core system with a
lightning fast SSD (230 MB/s). Okay, the first one used metacity (by
accident, I suppose), but that only explains about two seconds of the
difference.

But even if that is a local problem, boot speed keeps being a commonly
raised requirement, and it becomes even more important on low
performance devices like ARM boards.

Martin

[1] http://people.canonical.com/~pitti/bootcharts/donald-maverick-20100805-1.png
[2] http://people.canonical.com/~pitti/bootcharts/donald-maverick-20101004-1.png
-- 
Martin Pitt                        | http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)