On Mon, 16 Aug 2010 17:36:51 -0500, Brian Peterson 
<brian.peterson.ibm.m...@comcast.net> wrote:

>On Mon, 16 Aug 2010 15:26:54 -0700, Norman Hollander on DesertWiz wrote:
>
>>Very much agree, Brian.  You can never predict who will start up another
>>region or when.
>>Of course you may know when the phones start ringing with poor 
performance
>>to find
>>your system is paging excessively...
>>
>>zNorman
>
>I guess I would simply expect the system to do the best that it can with the
>resources it has been given.  I just think asking *me* to predict how many
>large pages my system will need for the life of the NEXT IPL is simply
>hopeless in terms of a strategy.
>
>Brian


While having the system (WLM) adjust the LFAREA size is a lofty goal, it is 
much easier said than done. Once large pages are allocated they are fixed and 
the system can not reclaim the frames until the application frees the virtual 
storage. So if the system needs more 4k frames it can not simply reduce the 
size of the LFAREA if the large frames are allocated. The LFAREA parameter is 
a means to let the OS know how much of your central storage you can afford 
to reserve for fixed frames. APAR OA3116 does provide some dynamic 
adjustment to 4K frames in the event the LFAREA frames are available. 

For the current exploiters of large pages you can estimate the LFAREA size by 
summing up how many large pages are needed for all your JAVA heaps and all 
your DB2 bufferpools (if running DB2 V10). APAR OA34024 will provide guidance 
on how to calculate the LFAREA size. 

I agree that in the long term we want to get to the point where the system 
self tunes the 1MB memory pool (LFAREA) and the 4KB memory pool. We are 
looking at this as a possible future enhancement.   

Elpida Tzortzatos
phone: (845) 435-3125
email: elp...@us.ibm.com

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to