I also tested (mail-)performance with a z/Linux and a Intel/Linux

z/Linux: 1024 MB RAM, 45 Mips min, 1 2064-CPU=250 Mips max, BogoMips about
770 (?)
x86/Linux : 256 MB RAM, PIII 700, about 1200 BogoMips (?)


There where a third x86/Linux with a P133 64MB RAM, which was configured to
relay mails with checking for virii (postfix+amavis). Of course a lame
system.

We send mails to this P133 in a loop from the 2 Testing Systems, which of
course were rejected after a while. 
So mails accumulate in outgoing queues...

RH7.1 PIII 700  : 130 mails in outgoing queue    :   Load of 42 !!!  (system
nearly locked)
RH7.1 z/Serie   : 4209 mails in outgoing queue   :   Load of 6  !!!  (system
laggy, but available)

Another day I let both Linux crunch a Seti@Home-Workunit (fast fourier
frequenzy-analysis):

RH7.1 PIII 700  : about 9 hours
RH7.1 z/Serie   : about 20 hours


I didnt have the time to perform the test accurate with exact protocols etc.
But I was surprised about that IO-Performance. And noBody ever stated, that
the Mainframe is a NumberCruncher...

any comments ? 

Frank Schwede


-----Urspr�ngliche Nachricht-----
Von: Moloko Monyepao [mailto:[EMAIL PROTECTED]]
Gesendet am: Dienstag, 19. M�rz 2002 10:25
An: [EMAIL PROTECTED]
Betreff: Bogomips S390(IFL) / Intel

I RedHat 7.2 (Server Installation). I am currentely testing sendmail
(Comparing the load handling between Linux on Intel and Linux on S390. I am
using an IBM z37 machine with one IFL and I gave my Linux LPar 80Pwt
(processor weighting on the IFL)
When the sever gets too busy start rejecting messages.  Just for outbount
mail and I also want to test for both Inbound/Outbound. What could be the
problem here. I will send any further information if it is needed. The
following is what I got from my mail administrator.

This is the BogoMips rating for the various machines:

Mainframe: 326.04 BogoMIPS (Mainframe)
Maggie (mail-spool, 500Mhz): 996.14  (Intel) 500meg (processor)

Higher is better, BogoMIPS should also be generic across platforms.

The mainframe is running very high load and is getting behind with
mail. Due to the high load, it starts to block connections which is no
good at all (might need to move it back). At this stage I am only
running outbound mail through the box, which is not really where the
load comes from. The real load comes from inbound mail (300-400
concurrent connections). I don't think this is going to work, unless you
can get MUCH faster processors. At least in the range of 2Ghz+ as in
Intel architecture.

Please assist
Moloko

Reply via email to