[2] The cluster's nodes are arrayed in five logical sub-clusters each with different memory and storage configurations designed to meet the varying needs of the researchers using Coates.
Coates nodes run Red Hat Enterprise Linux 5.6 (RHEL5.6) and use Portable Batch System Professional 10.4.6 (PBSPro 10.4.6) for resource and job management.
Each faculty partner always has ready access to the capacity he or she purchases and potentially to more computing power when the nodes of other investors are idle.
Unused, or opportunistic, cycles from Coates are made available to the TeraGrid and the Open Science Grid using Condor software.
The Coates cluster continues Purdue's practice of naming new supercomputers after notable figures in the university's computing history.