Description: Space presents a challenging environment for computing. Extended development times and radiation tolerance requirements leave hardware performance a decade or more behind the terrestrial state-of-the-art at the time of deployment. Additionally, once deployed, hardware changes are impractical, encouraging a trend towards increased software programmability. However, topside pressure from application advancements are forcing space-based platforms to improve throughput and latency while reducing power consumption. A popular approach to addressing the tension between these requirements is the heterogeneous processing architecture. By providing multiple hardware tools that optimally support a subset of the anticipated workload, a heterogeneous architecture can offer a diverse processing toolset to the application developer. However, programming these systems is extremely challenging because of variations in toolsets and data sharing interfaces. As a result, data sharing and dynamic workload scheduling across heterogeneous architectures is often suboptimal and hindered by poor scalability. Maxentric proposes to solve this problem with RUSH, a heterogeneous processing architecture with a unified programming model for rapid development. RUSH employs a rad-hard multicore processor as a host and an FPGA as an accelerator chip. The RUSH software layer unifies these architectures through an innovative programming model described in the proposal.

Benefits: The applicability of RUSH is not limited to NASA markets. The military and civilian satellite designers and operators stand to benefit from a highly-reconfigurable performance computation platform. Target markets include the space and nuclear computing platforms for diverse activities from real-time video processing, to software defined radio, to sensor processing.

RUSH will provide a high-performance, flexible, and programmable heterogeneous processing platform for space applications. This work has broad impact for space command and control systems. Reconfigurable high-performance systems with updates will enable deployed systems to maintain high levels of utility as processing or mission demands shift. The system is well qualified to enable software defined radio, real-time video processing, sensor analysis, and automated landing and will play a central role in research, development, and deployment of platforms for achieving these requirements.