On Mon, 22 Oct 2001, Bijoy Paul wrote:

 |This sounds something differenet from what Raju was
 |telling. So i made a google search and got two links
 |with contradictory information, 1 supports raju and
 |the other kingsly..

 |http://lithos.gat.com/docview/load_average.html
 |http://www.useforesite.com/uptime.shtml
 |
 |Also  if kingsly is right what does it mean by a load
 |average of 405 (as Sridhar mentioned) !!!!!!
 |
 |So how to come to a conclusion...????

My knowledge of this comes from what I either read on some website or what
someone told me on IRC a few years ago.. not exactly out a textbook/manual.

If you run john(the ripper) on a machine

1 instance Load = 1.00 CPU =100%
2 instances Load = 2.00 CPU=100%
3 instances Load = 3.00 CPU=100%

and when you run two instances of john ... the performance is down to
50%(each) ... and so a load average of over 1.00 definitely results in
system slowness, that lead me to believe that the explaination was
correct...

But going by the other theory ... it would be okay to have load averages
of over 1.00 ... but ofcourse they jobs would need to be small ones.. that
end quickly.. so the system slowness won't be noticed.

Maybe somebody can give us a link to some more authentic docs.

Kingsly a.k.a "Someone" :o)

PS:

I once noticed a load avg of 99 on a clients machine... with no processes
running ..but there was no system slowness either ... but after a reboot
it got back to normal numbers so I have no idea how that happened.

On my machine the load once shot to 13 when too many instances of
festival started due to a flood of messages on icq.. and the slowness was
really noticable on my p3-733 ..  infact I noticed the system becoming
dead slow before the deafening tts output hit my ears.



_______________________________________________
linux-india-help mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/linux-india-help

Reply via email to