Comparative Analysis of Present and Future Space-Grade Processors with Device Metrics

Abstract

Due to harsh and inaccessible operating environments, space computing presents many unique challenges and constraints that limit onboard computing performance. However, the increasing need for real-time sensor and autonomous processing, coupled with limited communication bandwidth with ground stations, is increasing onboard computing demands for next-generation space missions. Because currently available space-grade processors cannot satisfy this growing demand, research into various processors is conducted to ensure that potential new processors are based upon architectures that will best meet the computing needs of space missions. Device metrics are used to measure and compare the theoretical capabilities of processors based upon vendor-provided data and tools, enabling the study of large and diverse sets of architectures. Architectural tradeoffs are determined that can be considered when comparing or designing space-grade processors. Results demonstrate how onboard computing capabilities are increasing due to emerging architectures that support high levels of parallelism in terms of computational units, internal memories, and input/output resources; and that performance varies between applications, depending on the compute-intensive kernels used. Furthermore, the overheads incurred by radiation hardening are quantified and used to analyze low-power commercial-off-the-shelf processors for potential hardening and use in future space missions.

Comparative Benchmarking Analysis of Next-Generation Space Processors

“…Previous research has been conducted to evaluate the performance of radiation-hardened processors. The study in [13] investigated the capabilities of the RAD5545 and several other CPUand FPGA-based computing systems via performance metrics analysis. These metrics provide insight into performance characteristics such as computational density and memory bandwidth without requiring the device for analysis.…”

“…This system is also specifically designed for on-board processing applications, equipped with four RAD5500 TM Power Architecture processor cores to conduct computations in an efficiently parallel manner. This processor is capable of 5.6 GOPS (billions of operations per second), 3.7 GFLOPS (billions of floating-point operations per second), and up to 1398 DMIPS per core, for a total of 5592 DMIPS, at 466 MHz [13]. Its capability is aided by three levels of cache as well as the ability to interface with other devices via Serial RapidIO for high-speed communication [18].…”

“…The RAD5545 and HPSC processors, being radiation-hardened, have a significantly lower clock frequency than the facsimile platforms assessed. Previous performance studies for radiationhardened processors have employed frequency scaling as in [13]. This research proposes a hybrid approach to isolate and minimize scaling error by unifying two methods: underclocking and frequency scaling.…”

“…It was therefore decided that frequency-scaling methods to the RAD5545 projected frequency of 466 MHz consistent with[13] would be applied. The ODROID-C2 HPSC facsimile was much more straightforward to underclock via a software frequency governor integrated as a component of the processor driver.…”

Radiation-hardened MRAM-based LUT for non-volatile FPGA soft error mitigation with multi-node upset tolerance

“…Radiation-induced soft errors in nanometer-scale electronic circuits are of increasing concern in missioncritical space-based [1], high altitude [2], and terrestrial applications [3]. For instance, space missions which take place in a harsh environment in terms of cosmic radiation particles, temperature, and electromagnetic disturbances have grappled with radiation-induced upsets for many decades of continued device technology scaling [4].…”