The SU/sec advertised for those models indicates it is correct to call a
504 cpu "slower" than a 701 cpu, even if that distinction is created at
a level above the physical hardware. If it is true that at the hardware
level the processor clock cycle is identical for all processors, then
obviously that is not the only determinant of the effective cpu
execution speed. I would guess that the capacity restrictions described
force some desired percentage of clock cycles to be idle/no-work cycles,
reducing the effective execution speed of the processor. That does seem
like an ingenious approach for allowing modest upgrade paths in an age
where adding a full capacity cpu can be way too large of an increment.
JC Ewing
On 7/15/23 05:43, P H wrote:
All processing units (PUs), whether CPs zIIPs, SAPs, ICFs or IFLs have the SAME
cycle time (speed). It's the CAPACITY settings which are different. For
example, in case of the z16 the 701 is 100% then the relative capacity of a 601
is approx 66% of the 701, 501 is approx 41% of the 701 and 401 is approx 12% of
the 701
Regards
Parwez Hamid
________________________________
From: IBM Mainframe Discussion List <IBM-MAIN@LISTSERV.UA.EDU> on behalf of Jon
Perryman <jperr...@pacbell.net>
Sent: 15 July 2023 05:54
To: IBM-MAIN@LISTSERV.UA.EDU <IBM-MAIN@LISTSERV.UA.EDU>
Subject: Re: A question about CPU usage on z/OS
While each cpu in the 504 is slower than a 701 cpu, running 4 batch jobs at the
same time should reduce run time because each batch job expect reduced wait
because there is reduced competition for the CPU. However, you could be correct
if the 4 batch jobs are experiencing heavy I/O wait.
On Friday, July 14, 2023 at 06:05:24 PM PDT, Tom Brennan
<t...@tombrennansoftware.com> wrote:
On 7/14/2023 3:01 PM, Jon Perryman wrote:
> As for batch running slower at night after you went from 1 CPU to 4,
that doesn't make sense unless other things changed.
I'm thinking it could be as simple as say, going from a 701 to a 504.
The overall MIPS are bumped up, multi-task address spaces are happier,
but single threaded work can be left in the dust.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
--
Joel C. Ewing
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
