Combining Functional And Performance Debugging Of Parallel And Distributed Systems Based On Model-driven Monitoring

Abstract

In order to program paraller' and distributed systems efjciently, a systematic way ojl" dejning and understanding the complex behavior of process interactions in concurrent programs is needed. A proven method for understanding existing programs is event-driven monitoring which abstracts the program behavior to a sequence of events. These event traces are analyzed for debugging and tuning of the program.A new method is model-driven monitoring. By building a functional model of the parallel program describing the essential properties for debugging and performance evaluation, the deJinition of monitoring events is automated and carried out systematically. This integration of modeling and monitoring guarantees the same set of events in monitoring and modeling. It enables validation, i.e. functional debugging, of the program behavior by checking the event trace against the behavior represented in the model. I f a functional error occurs, the iillformation contained in the model allows the localization of the error in the program. However, strange performance behavior of a functionally correct program noticed during the event trace evaluation cannot easily be attached to specijic parts of the model or of the program. We solved this localization of pe~ormance errors by synchronizing event trace evaluation with event trace validation.

A Formal Framework For Specifying Design Methods

“…As these activities are usually related in different ways by tools and methods, we have decided to consider a concurrent view of these groups of activities (as illustrated in Figure 1). This means, for example, that performance prototypes can be developed both in the early stages of the development [8,15], and after the implementation [4].…”

“…So, the model allows for the hardware description to be constructed and analysed simultaneously with the software. This enables us to model SDMs where the hardware-dependent designs are derived from abstract (virtual) designs [13], and also those where the hardware characteristics are taken into account during the design, especially to reduce "performance design errors" [4].…”

A Formal Framework For Specifying Design Methods

“…As these activities are usually related in different ways by tools and methods, we have decided to consider a concurrent view of these groups of activities (as illustrated in Figure 1). This means, for example, that performance prototypes can be developed both in the early stages of the development (Ferscha and Johnson 1995, Justo 19961, and after the implementation (Dauphin 1994).…”

“…So, the model allows for the hardware description to be constructed and analysed simultaneously with the software. This enables us to model SDMs where the hardware-dependent designs are derived from abstract (virtual) designs (Jamieson 1987), and also those where the hardware characteristics are taken into account during the design, especially to reduce 'performance design errors' (Dauphin 1994).…”