On Tue, Jun 16, 2009 at 3:08 AM, Brock Pytlik<[email protected]> wrote:
> Peter Tribble wrote:
>>
>> I'm not sure I would call two and a half minutes on first boot all that
>> fast.
>> (That's on the sparc box I installed today, which has been slimmed down
>> a bit - one of my key minimization priorities now is not to reduce the
>> amount
>> of software delivered but the number of manifests, as that's a limiting
>> factor.
>> An opteron box is quicker, as you would expect, but you're still looking
>> at a
>> minute which is appreciable. And there's several seconds every boot as
>> well.)
>
> Is first boot really the use case you think we should be optimizing for? I'd
> like to think that for both users and machines, it's a comparatively rare
> event.

But a crucially important one. For users, it directly impacts on their initial
experience of the OS - twiddling your thumbs while the OS takes seemingly
forever to boot is very discouraging. For machines, think dynamic deployment
of zones - adding several minutes to the time it takes for a newly deployed
system to be available for use isn't very dynamic.

-- 
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss

Reply via email to