Modeling and Analysis of MAPE-K loop in Self Adaptive Systems using Petri Nets

摘要

Feedback loop plays a pivotal role in modeling of systems that have capability to adapt to new requirements during execution. These systems are categorized as self-adaptive systems (SAS) which can alter their working according to the inputs received from the environment. At present, we are surrounded by software systems that are either adaptive or self-adaptive. In both cases feedback loop has a major role in the adaptation process. Hence, reliable and efficient working of this loop is critical towards successful development of software systems that have ability to work with requirements that were not known at the time of development. Formal methods are mathematics of software and hardware systems, these methods include modeling languages and tools to model and analyze systems with use of concrete mathematical principles. Petri-Nets is a formal specification language that is used for analysis, modeling and testing of complex systems. In this paper we have presented an initial model of feedback loop using Petri-Nets. It has been observed that modeling and analysis of feedback loop using Petri-nets has been useful in verification of system at an abstract level and the generated model is free from deadlock and has capability to expand in future. This is an abstract model with limited inputs, invariants and constraints, this model will be enhanced for a complete SAS in future research.