From: Jonathan C. <jon...@gm...> - 2011-07-27 22:37:53
|
Hello! I'm a new OpenQRM user and have spent several days playing around with some very basic configurations. Rather than continue to bash my head against the wall I wanted to share my goal with the list and see if anyone has any idea of the best way to achieve them, or if they are even possible. Here is my scenario: I have a cluster of 16 48-core compute nodes (4x AMD Magny Cours procs per node). In a perfect world here is what I am looking to do: - Provide the ability for customer(s) to log into the portal - Customer makes a request like: "I want (4) 4-core systems running CentOS 6.0" - The appropriate KVM instances get deployed onto one or more available compute nodes - The customer gets their login information and gets to work - Repeat the above thing several times a day for different users Most of the examples I have seen involve deploying KVM or Xen instances on the OpenQRM system itself. That's not terribly helpful in this situation unfortunately. Any help is most appreciated and I am happy to provide more detail! - Jonathan |