Far be it for me to add flammable material to an already-raging forest
fire, but ...
Consider a z9 processor with CP, IFL, zAAP and zIIP processors ...
with "n" LPARs, all with both defined and reserved LPs in case someone
invokes CUOD
managed by IRD - which can and will (and does) fiddle with the number of
online LPs in the various LPARs.
The logical:physical ratio is not a constant, it's become a variable.
How do IFL, zAAP and zIIP factor into the logical:physical ratio?
Tom Puddicombe
Mainframe Performance & Capacity Planning
Computer Sciences Corporation
Registered Office: 2100 East Grand Avenue, El Segundo California 90245, USA
Registered in USA No: C-489-59
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
This is a PRIVATE message. If you are not the intended recipient, please
delete without copying and kindly advise us by e-mail of the mistake in
delivery.
NOTE: Regardless of content, this e-mail shall not operate to bind CSC to
any order or other contract unless pursuant to explicit written agreement
or government initiative expressly permitting the use of e-mail for such
purpose.
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Mark Zelden
<[EMAIL PROTECTED]
CHNA.COM> To
Sent by: IBM [email protected]
Mainframe cc
Discussion List
<[EMAIL PROTECTED] Subject
.EDU> Re: Logical to Physical CPU ratio
ROT (was Re: IXC454I messages)
10/11/2007 09:57
AM
Please respond to
IBM Mainframe
Discussion List
<[EMAIL PROTECTED]
.EDU>
On Wed, 10 Oct 2007 23:04:24 +0000, Ted MacNEIL <[EMAIL PROTECTED]> wrote:
>>I thought the rule of thumb for Logical to Physical CPU's was a 2:1
ratio.
>
>As always, it depends.
>The CP:LP ratio is dependent on the activity of the LPARs. NOT on the
number of LP's.
>
>I've gone 3:1 with low activity images.
We've used 4:1 (or higher) on our "penalty box"...
All in the archives....
The old ROT was 2:1 and basically still applies. It is also still
documented in
the RMF Performance Management guide (Appendix A). I remember seeing
something about 3:1 in some z990 white paper, but it may have been an
IBM "internal use". The best advise is not to define more than you need
to do the work.
Bottom line... if it works and performance is acceptable, it's ok. If you
have a z9 with 2 processors and have 5 LPARs all with 2 LPs,
(5:1 ratio), whats the harm if the box never hits capacity (some people
are not comfortable defining an LPAR with only 1 CP).
Look on the techdocs website for a white paper about performance
considerations when moving to fewer faster CPUs. IIRC, there is a
benchmark in done in there comparing CICS and batch response times
between a 9672 and a z900 with a 10:1 LP/CP ratio (the z900 was
better). A little dated.. but think of how much faster a z9 engine is
than a z900!
HTH,
Mark
--
Mark Zelden
Sr. Software and Systems Architect - z/OS Team Lead
Zurich North America / Farmers Insurance Group - ZFUS G-ITO
mailto:[EMAIL PROTECTED]
z/OS and OS390 expert at http://searchDataCenter.com/ateExperts/
Systems Programming expert at http://expertanswercenter.techtarget.com/
Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
