BOINC + Cluster
I want to build a 'cluster' of computers running BOINC. I use the term cluster in the following context: Each computer is a unique copy of an OS (maybe Ubuntu or other) that has its own separate connection to BOINC and determines its own workload.
I want to use the BOINC system as it was intended. Each computer determines its processing ability, then communicates with projects to get work units. I don't want a central server that grabs BOINC work and then distributes it across the cluster. [eg not this: https://wiki.ubuntu.com/EasyUbuntuCl...edClusterGuide]
To simplify management I want to update only one computer to update. I want to have a master image or 'server' that is imaged by the 'nodes'. The hardware will be different across the cluster. I think I can use DRBL (Diskless Remote Boot in Linux)http://drbl.sourceforge.net/ but I'm not sure. However each node must have its own individual storage area of BOINC storage and configuration.
To reiterate: I want to build a 'cluster' of individual notes that are all performing BOINC processing, each with it's own individual instance of BOINC.
I think this is the proper way to properly use BOINC while still building a cluster. If you can point me in the right direction that would me amazing!