Re: 57##-SS1 option 42 HA Journal Performance

[Vern Hamberg <vhamberg@xxxxxxxxxxx>]

Charles

I was going to say the same - Rick Turner mentions this in his batch 
performance articles, still available in IBM's knowledge base. Just 
journaling a PF and then wrapping the process in commitment control, it 
does a certain measure of caching the writes to the journal receivers. 
Option 42 for free!

This PDF has a lot of info - 
ftp://service.boulder.ibm.com/as400/web/benchmark/batprf.pdf - and page 
13 has this paragraph -

"commitment control already performs some journal caching"

Vern

Charles Wilt wrote:
> On Thu, Apr 1, 2010 at 2:35 PM,  <rob@xxxxxxxxx> wrote:
>   
> > Got a lot more bang for the buck with the option 42 versus the SSD's
> > ***IF*** you are journalling.
> >     
>
> AND not using commitment control.
>
> If you used commitment control, and batched your commits, option 42
> wouldn't get you much more.
>
> Charles
>