AFAIK, ASM has always [1] worked this way. 

Page slots seem to be allocated in a "round-robin" fashion so that
Slot usage (not percentage) should be roughly equal across all local page data 
sets.

[1] This may not have been true for some of the early releases of MVS (think 
pre-ESA).

HTH

<snip>
The LPAR currently has 19 GB of real storage, quite a bit more than when I 
started attacking this problem last year. 

This observation comes from the gut. Or from the nose. (Is there a single nerve 
that connects them?): the growth in aux usage is out of proportion to any 
growth in LPAR usage. The pattern seemed to be that every page data set I added 
would get filled up to more or less the same percentage within more or less the 
same time period. So maybe 8 is not enough? What would happen with 16? With 32? 

I think that something is wrong. 
</snip>

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

Reply via email to