On reducing test time and meeting deadlines in real-time systems

Abstract

The authors investigate methods for reducing the time to test the suspected faulty units in a real-time system. They provide a clue to the tester about possible faulty locations within each unit. Hence only a fraction of the resources within a unit need to be tested. This is accomplished by keeping track of the resources used, called the hardware utilization vector (HUV), when a program is executing. The authors demonstrate the average reduction in test time of cache and memory subsystems and improvement in meeting deadlines under the presence of faults.<>