Peter,

2 points:


  1.  The way Speed Recovery Boost (SRB) or Instant Recovery (IR) was 
'announced' and not explained in detail, I wanted to make sure folks didn't get 
the impression that somehow the element of 'speed' was achived by giving a 
'boost' to the cycle time. As you rightly point out it has nothing to do with 
the cycle time. As we all well know or should know that sub-capacity and full 
capacity CPs have the same cycle time:-)
  2.  As regading the use of the word 'speed', you have pointed to one element 
of it and I pointed out the other one.
     *   As you state the 'speed', is going from sub-capacity to full capacity 
CP
     *   My point about 'speed' is that with the additional capacity provided 
by SRB i.e either by sub-capacity to full capacity CPs or CP workoad being 
dispatched to zIIPs will help bring up/down the OS and sub-systems quicker in 
the LPAR which gets SRB.

PS: It hasn't helped when some documentation states SRB or IR and some of the 
announcement material wasn't really very clear what was meant by 'speed' đŸ™‚


Regards

Parwez Hamid​

________________________________
From: IBM Mainframe Discussion List <[email protected]> on behalf of 
Peter Relson <[email protected]>
Sent: 29 November 2019 15:26
To: [email protected] <[email protected]>
Subject: Re: OOCoD experiences?

Parwez Hamid​ wrote
<snip>
Some of the documentation states 'speed'. Remember, this refers to the
'speed' of bringing down/up of the OS and sub-systems. This is done by
ACTIVATING ADDITIONAL capacity and not by changing the GHz of the
processor.
</snip>

If you are referring to the "speed boost", that is not correct. The
"speed" within "speed boost" does not refer to the speed of bringing
down/up anything. It does not refer to activating additional capacity
(such activation is part of the "zIIP boost" AKA the "zIIP capacity
boost"). It does not refer to (and does not have anything to do with) the
cycle frequency, which you may think of as the GHz of the processor.  It
refers to the effective speed of the processor by moving it from
sub-capacity to full capacity for the duration of the boost.

The fact that the cycle frequency of the machine does not change to
accomplish the speed boost is a reason that the boost is not called
"frequency boost".

Peter Relson
z/OS Core Technology Design


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to