Soar-Longevity: A Sustainable Autonomic Architecture for Organically Reconfigurable Computing Systems

Soar Technology proposes the Soar-Longevity autonomic architecture for self-adapting sustainable performance. Soar-Longevity utilizes a two-tiered approach to synergize lower-level organic computing elements and upper-level cognitive supervisory elements. Soar Technology’s Cognitive Layer monitors the organic computing elements with respect to mission performance specifications. The Cognitive Layer also leverages the organic computing elements' reporting capabilities to assist in fault isolation, resolution, and avoidance of undesirable emergent behaviors at the system level. The Organic Layer resides beneath it and is composed of Field Programmable Gate Arrays (FPGAs) containing Functional Elements (FEs) and Autonomic Elements (AEs). Thus, each FPGA is imparted with an independent capability to self-monitor and self-repair using the techniques developed and demonstrated by the University of Central Florida (UCF). Soar Technology, Inc., teaming with researchers from UCF, will design and develop the integrated architecture. Soar-Longevity will be capable of maintaining mission requirements through monitoring and management of functionality to independently sustainable FPGAs, optionally including the FPGAs that comprise components of the Cognitive Layer itself.