PolyCluster is a high-performance cluster funded by NSF through
MRI-0619911. It contains 16 nodes, each with 4 processors.
This is a community resource, available for Cal Poly faculty and
associated students who need substantial computing resources.
The machine became functional in April, 2007.
If you use the machine for results that get published, you need to add the following sentence to your publications: "This work was supported by NSF MRI-0619911 to Diana Franklin."
Several projects have begun using the machine.
Evidence showing that some instructions are more important than others. We have a compiler that identifies instructions to protect. If errors are randomly inserted anywhere in the code, catastrophic failures (failures causing early termination) occur very often. If we protect a subset of the instructions, even with the same rate of error insertion, the number of catastrophic errors is reduced dramatically.