An integrated architecture for distributed real-time systems analysis
Jeffrey J. P. Tsai, Stephen J.H. Yang, A. Liu · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 1996
Analyzing distributed real-time software systems is considered difficult because of imposed timing constraints and non-deterministic behavior. These two factors inhibit the reproduction of an error without interfering with program execution, where error reproduction is the corner-stone of conventional software analysis techniques. Statistical evidence indicates that testing and debugging represent approximately 50% of the cost of new system development. In the case of distributed real-time software, the percentage is as high as 70% because the errors are {open_quotes}immune{close_quotes} to conventional debugging aids. This paper presents an integrated architecture which supports monitoring, visualization, and debugging methods for analyzing the dynamic behavior of distributed real-time systems. Based on our approach, an execution environment will be developed to simulate the program execution and data collection activities. Different levels of logical views are then reconstructed from these monitored information. A dynamic visualization and timing analysis method will then be used to study and analyze the timing behavior of distributed real-time software systems.