I concur with Shane, We ran our Development Sysplexes this way for 8 years (we got extra ICF last year).
ICF lpar on one box with all the structures. A CF lpar sharing GP engines on the other box (for those few times when we wanted to shutdown the ICF lpar). Never had a problem. Regards Bruce Hewson On Thu, 13 Jul 2006 08:30:38 +1000, Shane <[EMAIL PROTECTED]> wrote: >Mmmmm ... where to start. > <snip> > >Now for some heresy; one ICF should suffice. The mean time to failure of >an engine in this scenario is probably measured in units of centuries. >And what is the likelihood of the ICF going away, and the rest of the >machine staying up ??? >ZERO - or a very close approximation. > >To satisfy "backup" requirements, maybe create a CF LPAR as a fallback, >and force all the structures to reside in the ICF by default. Causes >some interference with the CFs having to talk to each other, but should >work o.k. Personally I wouldn't bother if I could get the boss to agree >to the "exposure". <snip> >Shane ... > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
