Non-Markovian analysis for model driven engineering of real-time software

Abstract: Quantitative evaluation of models with stochastic timings can decisively support schedulability analysis and performance engineering of real-time concurrent systems. These tasks require modeling formalisms and solution techniques that can encompass stochastic temporal parameters firmly constrained within a bounded support, thus breaking the limits of Markovian approaches. The problem is further exacerbated by the need to represent suspension of timers, which results from common patterns of real-time programming. This poses relevant challenges both in the theoretical development of non-Markovian solution techniques and in their practical integration within a viable tailoring of industrial processes. We address both issues by extending a method for transient analysis of non-Markovian models to encompass suspension of timers. The solution technique addresses models that include timers with bounded and deterministic support, which are essential to represent synchronous task releases, timeouts, offsets, jitters, and computations constrained by a Best Case Execution Time (BCET) and a Worst Case Execution Time (WCET). As a notable trait, the theory of analysis is amenable to the integration within a Model Driven Development (MDD) approach, providing specific evaluation capabilities in support of performance engineering without disrupting the flow of design and documentation of the consolidated practice.

Proceedings of ICPE, pp. 113-124, ACM, 2013
Stochastic ProcessesPerformance ModelsApplications



copy bib | save bib | save pdf | go to publisher

🏠 Home