On Wed, 14 Apr 2010 11:20:33 -0300, ITURIEL DO NASCIMENTO NETO <[email protected]> wrote:
>Mr Sinram, > >Thanks for your feed back. Specifying EXTENDED_DATA=YES did the trick. > >Actually, i'm trying to find out the most idle system to direct work to. >To do it i'm not just only evalueting Unused Service Units, which >sometimes >presents value 0 even when the system is not full. > >I'm evalueting the most idle system with the following calculation : > >Busy% = (Sum all Service units in the last 60 seconds) / (Num of >Processors * Cpu Speed per second * 60) > >This formula seems rigth to me, but Busy% is not even close of what RMF >shows. >I know that i'm comparing apples and oranges, but they're both round. > I would argue that a busy rate is not a good metric to base routing decisions upon. The LPAR sizes and underlying hardware may be very different for the members of a Sysplex such that %Busy does not allow to judge which system is best suitable for additional work. An absolute measure, such as the raw service units returned by IWMWSYSQ, however, does allow comparing heterogeneous LPARs. It also provides you with an understanding of the displaceable capacity at a given importance level. Kind regards. Horst Sinram ,IBM z/OS DFSMSrmm Architecture, z/OS Capacity Management ---------------------------------------------------------------------- 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
