Performance and dependability modeling with Möbius

  • Authors:
  • Shravan Gaonkar;Ken Keefe;Ruth Lamprecht;Eric Rozier;Peter Kemper;William H. Sanders

  • Affiliations:
  • University of Illinois, Urbana, IL;University of Illinois, Urbana, IL;College of William and Mary, Williamsburg, VA;University of Illinois, Urbana, IL;College of William and Mary, Williamsburg, VA;University of Illinois, Urbana, IL

  • Venue:
  • ACM SIGMETRICS Performance Evaluation Review
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Möbius is a multi-paradigm multi-solution framework to describe and analyze stochastic models of discrete-event dynamic systems. Möbius is widely used in academia and industry for the performance and dependability assessment of technical systems. It comes with a design of experiments as well as automated support for distributing a series of simulation experiments over a network to support the exploration of design spaces for real-world applications. In addition to that, the Möbius simulator interfaces with Traviando, a separate trace analyzer and visualizer that helps to investigate the details of a complex model for validation, verification, and debugging purposes. In this paper, we outline the development of a multi-formalism model of a Lustre-like file system, the analysis of its detailed simulated behavior, and the results obtained from a simulation study.