Are you sure? What ratio are you using for your sizing estimates?

I have not read or found anything to refute my prior estimate, going back to 
MVS/ESA
(ignoring ESTOR paging and SWAP which are now obsolete, and possibly paging 
above the bar).

Yes the ASM algorithms have been updated to handle larger real storage sizes 
with less overhead, primarily by invoking the routines 
less frequently under low stress, however, the actual paging routines 
themselves have not been altered significantly with the exception of 
SUSPEND/RESUME removal and PAV support. 

I might  budge on the PPS/Page dataset, but that's about it. 
Channel/Device/Control unit issues are smaller due to the advent of 
ESCON/FICON/PAV.

A given is that particular set of workloads in an LPAR with X amount of real 
storage will produce a paging rate of Y.
If the workload(virtual storage wise) is increased without a corresponding 
increase in real storage the paging rate will increase as the square of the 
ratio.
Likewise, if real storage is removed without changing the workload, the same 
behavior will occur.


<snip>

For most installations your virtual-to-real ratio is way out of date.

Martin Packer,
zChampion, Principal Systems Investigator, Worldwide Banking Center of 
Excellence, IBM
</snip>

<snip>
The "optimum" VSTOR config is 2:1 virtual to real. I personally use 3:1 (RSTOR 
*3). This is a trade off between page rate and devices allocated.

I.e. the sum of all working sets for the LPAR(in frames)  is up to 3 times the 
number of real storage frames available.
This paging rate goes up *VERY VERY* quickly after 3:1 IIRC, this is a SQUARE 
function.
For a given original page rate:
To go from 2:1 virtual/real to 3:1 virtual/real generates the ratio (9/4 * 
original page rate) 2.25 times To go from 2:1 virtual/real to 4:1 virtual/real 
generates the ratio (16/4
* original page rate) 4 times the original page rate.
</snip>

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

Reply via email to