My tangent on this wasn't brute cpu power, it is on the cost of such power.

You know as well as I do, that it is very hard to justify mainframes
just on hardware cost.  But when you consider the manpower cost and use
of the "white" area along with environmentals, there is justification.
And if you can throw in the cost of software licenses, so much the better.

Of course if the cost of manpower, isn't a concern, forget about all
computers and get a building full of clerks with pencils <G>.

And as I've said in one of the past posts on this thread, that I've
always been of the opinion that CPU intensive workloads really were not
suitable for Linux/390.  But that guy running 7 images on 9 IFL engines
(that to me, fits the description of CPU intensive), does have me wondering.

For the last year, I have talked up putting Linux images on the
mainframe.  And if they become too resource hungry, we can move them to
another platform, but with the 7 images on 9 IFL engines, apparently
being justifiable, I've backed off on that line of justification.

Tom Duerbusch
THD Consulting

John Summerfield wrote:

On Wed, 18 Jun 2003, Tom Duerbusch wrote:



The part of offloading cycles to a cheaper platform, is that we would
be offloading to a more expensive platform (intel).  Not that the Intel
box isn't cheap, but the economic reasons for server consolidations is
to get away from these "cheap" boxes.




What's cheap and what's expensive depends on the workload. People aren't
using IA32 CPUs to make super computers because that path's dear. They
do it because it's the cheapest way to get the CPU power.



Until a few months ago, I've had the impression that putting cpu type
loads on the mainframe wasn't economical compared to putting the same
loads on Intel or Sun platforms.

But then I start hearing about some other sites, one that had 7 Linux
images in LPAR mode, using 9 processors.  Apparently, it was
economically justifiable.  I still don't understand how.  But it did
open my eyes to "run the numbers" instead of throwing it out just based
on an outdated "rule of thumb".




Okay, run these numbers:
[EMAIL PROTECTED]:~$ time bin/bm.perl

real    0m3.247s
user    0m3.160s
sys     0m0.020s
[EMAIL PROTECTED]:~$ cat /proc/cpuinfo
processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 6
model           : 4
model name      : AMD Athlon(tm) Processor
stepping        : 4
cpu MHz         : 1410.211
cache size      : 256 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge
mca cmov pat pse36 mmx fxsr syscall mmxext 3dnowext 3dnow
bogomips        : 2811.49

[EMAIL PROTECTED]:~$ cat bin/bm.perl
#!/usr/bin/perl
#use integer;
$i = 0;
while ($i < 1000)
       {
               $j = 0;
               while ($j < 10000)
                       {
                               ++$j;
                       }
               ++$i;
       }

[EMAIL PROTECTED]:~$


This is on a cheap box bought on price. Well, it did have to be an Athlon, and I did want an Asus motherboard.

If you want a longer comparison, here's one with "while ($i < 1000)"
[EMAIL PROTECTED]:~$ time bin/bm.perl

real    0m31.986s
user    0m31.610s
sys     0m0.100s
[EMAIL PROTECTED]:~$

See you our CPU stacks up against mine.




--



Cheers John.

Join the "Linux Support by Small Businesses" list at
http://mail.computerdatasafe.com.au/mailman/listinfo/lssb



Reply via email to