To-Do List

Potential projects / tasks:

1. Telnet access to iSCB (to power-on/power-off/monitor nodes via the chassis)
2. Telnet access to PDUs (to monitor current power usage)
3. Powerman / conman integration (or whatever those handy LLNL utilities were...)
4. Cluster monitoring via Ganglia + web interface to view at a glance
5. Cluster monitoring via  Vampir. Possible student projects: Cluster usage visualization and performance analysis.
6. Research running VIRTUAL MACHINES on the cluster nodes (added flexibility for other applications / other research projects?)
7. GPU nodes
1. Order parts / complete build
2. Net-boot like other cluster nodes
3. Separate SLURM partition (only nodes with GPUs, plus that node does NOT have Inifiniband)
4. StarPU task programming library for hybrid architectures (http://starpu.gforge.inria.fr/#overview) could lead to more student projects.

Separate Project: LittleFe + XCAT installation  --- Suitable for INDEPENDENT STUDY PROJECT?

1. Do a test installation
2. Document document document
3. Try to script the entire setup!  Ideal deployment would be:
1. Install CentOS 7.1 (minimal?) with config settings of A, B, C
2. Run our EZ-configuration script
3. Reboot
4. Done!
4. PUBLICIZE PUBLICIZE PUBLICIZE