Software Reliability Engineering Case Study

摘要

Handling, identifying, and correcting faults are significant concerns for thesoftware manager because (1) the presence of faults in the operational software can put human life and mission success at risk in a safety critical application and (2) the entire software reliability process is expensive. Designing an effective Software Reliability Engineering (SRE) process is one method to increase reliability and reduce costs. This thesis describes a process that is being implemented at Marine Corps Tactical System Support Activity (MCTSSA), using the Schneidewind Reliability Model and the SRE process described in the American Institute of Aeronautics and Astronautics Recommended Practice in Software Reliability. In addition to applying thc SRE process to single node systems, its applicability to multi-node LAN-based distributed systems is explored. Each of the SRE steps is discussed, with practical examples provided, as they would apply to a testing facility. Special attention is directed to data collection methodologies and the application of model results. In addition, a handbook and training plan are provided for use by MCTSSA during the transition to the SRE process.