lkml.org 
[lkml]   [1996]   [Jun]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: real kernel bloat
   Date: Tue, 25 Jun 1996 23:02:42 +1200 (NZST)
From: "J. Sean Connell" <ankh@canuck.gen.nz>

No, and since this machine is about a month away from being our #1
machine, we're not going to be, either.

I can understand the concern. It's nice to be able to pass the buck
to a paid support center and tell your users "I have the engineer on
it now" when the machine tosses cookies. But we'll have that for
SparcLinux some day.

Not that I think badly of SparcLinux; it's just that when you have
a machine whose load average routinely goes above twenty (some days
above 40), you kinda wanna have a known good and rock-solid OS
behind it.

Load of 40? Try a load of 300/400 with 200 of those processes being
crashme invocations and the rest being parallel builds, this is on an
SMP 4 processor SS10 with 130MHZ HyperSparcs, and yes it runs
SparcLinux.

Meanwhile I can bring Solaris2.x down in under 30 seconds flat on any
sun4c/sun4m out there with very little effort....
I can't wait until there is a proof of concept SparcLinux box out
there doing "real work", and I think there will be soon. Anyone will
agree with me who has ever dealt with Sun support when they have hit a
serious bug ;-)

Later,
David S. Miller
dm@sgi.com


\
 
 \ /
  Last update: 2005-03-22 13:37    [from the cache]
©2003-2011 Jasper Spaans