I appreciate everyone's input. We are going to try to implement the VSWITCH.
Thanks Gene Gene Walters System Programmer WV Dept of Administration - OT 304-558-5914 ext 8902 Fax 304-558-1351 >>> [EMAIL PROTECTED] 5/14/2007 1:42:51 PM >>> > > If it were me, I'd probably go with the VSWITCH. Heartbeat packets > > really don't exploit the best parts of real hipersockets (high volume > > bulk data transfer) much, and the ability to separate stuff easily later > > is a big architectural win in my book. > > > Makes sense, particularly in our case where we have redundant OSA cards. Although since real hipersockets run over the system memory backplane, you also have multiple buses to work with there, so you get built-in redundancy. If that memory bus isn't working, then you're not doing much with the machine anyway, and it's always built in, so you don't have to worry about it at DR sites. I'd still do the VSWITCH, though. IMHO, it's a more flexible option. ---------------------------------------------------------------------- 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 ---------------------------------------------------------------------- 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
