>>> On Sun, Feb 24, 2008 at  9:24 AM, in message
<[EMAIL PROTECTED]>, Pieter Harder
<[EMAIL PROTECTED]> wrote: 
-snip-

>> My further understanding is that CMMA should not be used.
> Why? What is broken?

Pretty much the whole thing.  On a too-frequent basis it would go pathological 
and hurt you more than help you.  According to Martin Schwidefsky, however, 
about a week ago, he's changed his opinion on CMMA.  He gave me this URL to 
look at:
http://www.vm.ibm.com/perf/reports/zvm/html/530cmm.html  Look it over and make 
sure you meet the maintenance requirements.  Since I'm not a performance 
expert, I can't really assess the new-and-improved CMMA myself.  I have to 
depend on other people for that.

-snip-
> For SAP CMM-1 doesn't really help. SAP basically works on a 
> grab-everything-there-is-and-don't-release-it strategy.

Does SAP also touch every page it allocates on a frequent basis?  If not, then 
CMM-1 should still be able to help.

> I asked on the list 
> whether anybody had any experience or data on SAP and CMM-1, but got no 
> replies. Not few replies, but *none*. So I seem to be in virgin territory 
> there.

I'm on occasion working with a customer that is going to be doing this as well. 
 Our original advice was use CMM-1.  I may need to revisit that with them, but 
I haven't had much contact with them for a while.

> For now CMMA seems a lot easier to do (far less moving parts involved), and 
> I have all the requirements. 

Not sure how many moving parts you think CMM-1 has.  As far as I can see, it's 
loading the kernel module and echoing some values into /proc/sys/vm/cmm_*


Mark Post

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to