COTS-BASED FAULT TOLERANCE IN DEEP SPACE: A CASE STUDY ON IEEE 1394 APPLICATION
Ann T. Tai, Savio N. Chau, L. Alkalai · International Journal of Reliability Quality and Safety Engineering · 2002
Among the COTS applications in the X2000 architecture for deep-space missions, the use of commercial bus standards is the highest-payoff COTS application since a bus interface has a global impact on system cost and capability. While COTS bus standards enable significant cost reductions, it is a great challenge for us to deliver a highly-reliable long-term survivable system employing COTS standards that are not developed for mission-critical applications. The essence of our solution to the problem is to exploit the pertinent standard features of a COTS product to circumvent its shortcomings, including those standard features that are not originally designed for reliability purpose. In this paper, we discuss our experiences and findings on the design and assessment of an IEEE 1394 compliant fault-tolerant bus architecture. We first derive and qualitatively analyze a "stack-tree topology" that not only complies with IEEE 1394 but also enables the implementation of a fault-tolerant bus architecture without node redundancy. We then present a quantitative evaluation that demonstrates significant reliability improvement resulting from the COTS-based fault tolerance. We also describe the benefits from using COTS intellectual properties for hardware implementation of the bus architecture and discuss our general approach to COTS-based fault tolerance.