An architecture-driven software mobility framework

  • Authors:

  • Sam Malek;George Edwards;Yuriy Brun;Hossein Tajalli;Joshua Garcia;Ivo Krka;Nenad Medvidovic;Marija Mikic-Rakic;Gaurav S. Sukhatme

  • Affiliations:

  • Department of Computer Science, George Mason University, Fairfax, VA, USA;Computer Science Department, University of Southern California, Los Angeles, CA, USA;Computer Science & Engineering, University of Washington, Seattle, WA, USA;Computer Science Department, University of Southern California, Los Angeles, CA, USA;Computer Science Department, University of Southern California, Los Angeles, CA, USA;Computer Science Department, University of Southern California, Los Angeles, CA, USA;Computer Science Department, University of Southern California, Los Angeles, CA, USA;Google Inc., 1333 2nd Street, Santa Monica, CA, USA;Computer Science Department, University of Southern California, Los Angeles, CA, USA

  • Venue:
  • Journal of Systems and Software
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Software architecture has been shown to provide an appropriate level of granularity for assessing a software system's quality attributes (e.g., performance and dependability). Similarly, previous research has adopted an architecture-centric approach to reasoning about and managing the run-time adaptation of software systems. For mobile and pervasive software systems, which are known to be innately dynamic and unpredictable, the ability to assess a system's quality attributes and manage its dynamic run-time behavior is especially important. In the past, researchers have argued that a software architecture-based approach can be instrumental in facilitating mobile computing. In this paper, we present an integrated architecture-driven framework for modeling, analysis, implementation, deployment, and run-time migration of software systems executing on distributed, mobile, heterogeneous computing platforms. In particular, we describe the framework's support for dealing with the challenges posed by both logical and physical mobility. We also provide an overview of our experience with applying the framework to a family of distributed mobile robotics systems. This experience has verified our envisioned benefits of the approach, and has helped us to identify several avenues of future work.