On 1/16/09 9:04 AM, "Gentry, Stephen" <[email protected]> wrote:
> Let's take this in a different direction for a moment (a.k.a. OT) > Disregard the VDISK usage for a moment. The swap disk will be on a > mini-disk. What are the advantages (or disadvantages) of IPL'ing the > boot disk in the directory vs IPL'ing cms, doing some stuff in a profile > exec and then IPL'ing the boot disk (from the profile exec)? > I'm trying to keep storage usage "lean and mean" so IPL'ing CMS seems to > add an extra layer. Given that CMS is a shared segment, only one copy ends up in real storage, so you're really not buying all that much storage savings. You save a few cycles, but give up a lot of flexibility and configurability by removing the ability to have sophisticated REXX logic as part of your startup (you probably could do similar stuff inside the Linux guest, but REXX and CMS have a lot of interfaces and knowledge about the VM environment that have not yet been exposed to Linux. Having the COMMAND stuff in the CP directory does provide some configuration capability, but it isn't able to do conditional stuff. If you're concerned about disk space, put the virtual-machine specific stuff in a SFS directory where you use only the blocks that the specifications actually take up. If you're concerned about commonality, the suggestion of using a common PROFILE EXEC on a shared minidisk that calls a userid-specific EXEC for individual virtual machines is a time-tested and good way to do that. Short version: I can't really see a compelling advantage for putting everything in the CP directory and skipping the CMS IPL. It saves a few cycles at the expense of a lot of command-and-control benefits. -- db ---------------------------------------------------------------------- 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
