(non)importance of loadaverages

From: bert hubert (ahu@ds9a.nl)
Date: Sat Nov 11 2000 - 07:23:21 EST


On Fri, Nov 10, 2000 at 04:03:55PM -0600, Neil W Rickert wrote:
> "Jeff V. Merkey" <jmerkey@timpanogas.org> wrote:
>
> >The problem of dropping connections on 2.4 was related to the O RefuseLA
> >settings. The defaults in the RedHat, Suse, and OpenLinux RPMs are
> >clearly set too low for modern Linux kernels. You may want them cranked
> >up to 100 or something if you want sendmail to always work.
>
> If a modern Linux kernel requires high load average defaults, I will
> stop using Linux.

The importance people attach to loadaverages continues to amaze me. Two
systems doing the same work can have wildly different loadaverages. If I
code a big statemachine with lots of poll() interfaces, my loadaverage will
not get a lot higher then 1.

Should I forego writing a statemachine and use pthreads or fork(), the same
amount of work will keep lots of different processes busy and raise my
loadaverage wildly.

Do you now state that the second situation is somehow 'worse'?

Feel free however to stop using Linux. Or to quote the document Al refered
to 'See figure 1'.

Kind regards,

bert hubert

-- 
PowerDNS                     Versatile DNS Services  
Trilab                       The Technology People   
'SYN! .. SYN|ACK! .. ACK!' - the mating call of the internet
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed Nov 15 2000 - 21:00:19 EST