>This has always puzzled me, as it puts an inherent limit on CPU
capability.  I won't take very many "50% capacity improvements" to get to
zero...

I never claimed to be a math whiz, but wouldn't ie be impossible to reach zero 
my dividing any number repeately by 50%?  Granted, you can end up with a lot of 
numbers to the right of the decimal point, and the result gets ahrd to 
understand - worse: is counter-intuitive. Maybe this method was an attempt to 
relate to transaction speed?  When we get to zero, we've reached the end of 
processor speed improvements - unless someone develops a processor that 
provides results before the program runs!  Is these a target G.A. date for that 
yet?  Hmmm that be a REAL System z, zero down time, zero processing time!

Mike Walter


----- Original Message -----
From: "Alan Altmark" [EMAIL PROTECTED]
Sent: 05/01/2007 04:57 PM
To: [email protected]
Subject: Re: QUERY CAPABILITY question



On Tuesday, 05/01/2007 at 03:03 EST, Brian Nielsen
<[EMAIL PROTECTED]> wrote:
> The z/Architecture Principles of Operations describes the information
> returned by the STSI instruction from which the above values are
> obtained.  In particular it states (pg 10-117) that "a lower value
> indicates a proportionally higher CPU capacity."

This has always puzzled me, as it puts an inherent limit on CPU
capability.  I won't take very many "50% capacity improvements" to get to
zero which, I suppose, represents infinite capacity and will SOMEONE bring
me my pills!  I shouldn't worry so much....  hmmm....what will a cell
processor have?!?  <gaaaack!>

> The definition of which engines are considered PRIMARY and which are
> SECONDARY is not spelled out anywhere that I can find.  My expectation
was
> that the engines in use by the LPAR would be defined as PRIMARY.  Since
> the IFL's should be faster than the non-full-speed CP's I expected the
> PRIMARY to be faster, but the output from Q CAPABILITY seems to indicate
> the SECONDARY CPUs are faster.  This leads me to conclude that the CP's
> are (always?) PRIMARY regardless of what the engines the LPAR uses.  Is
> that correct?  If not, what is?

We should have labeled those "CPs" (primary) and "Others" (secondary). The
STSI information we are looking at is at the configuration (CEC) level,
not the partition.  In your case, yes, the secondaries are faster.  IFLs,
CFs, and zAAPs (and now zIIPs) run at full speed (smaller number).  Your
CP's are running at "subcapacity".

Alan Altmark
z/VM Development
IBM Endicott


 
The information contained in this e-mail and any accompanying documents may 
contain information that is confidential or otherwise protected from 
disclosure. If you are not the intended recipient of this message, or if this 
message has been addressed to you in error, please immediately alert the sender 
by reply e-mail and then delete this message, including any attachments. Any 
dissemination, distribution or other use of the contents of this message by 
anyone other than the intended recipient 
is strictly prohibited.

Reply via email to