A unified framework for design and performance analysis of distributed systems
- 27 November 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10872191,p. 109-118
- https://doi.org/10.1109/ipds.1998.707714
Abstract
Although it becomes more and more accepted that it is important to take into account performance and dependability issues throughout the design process of distributed systems, design languages and quantitative modelling formalisms are still separate worlds. This is largely due to the fact that their requirements are very different: design languages must offer ways to easily build structured comprehensible system specifications, while quantitative models should allow for efficient and accurate analysis. Moreover, system designers and performance modellers generally have a very different background. Aiming to bring these worlds together, we introduce a design language which includes quantitative properties of systems, and we show how specifications in this language can automatically be translated to a number of popular formalisms for quantitative distributed system modelling. The most appropriate performance modelling formalism to translate to is chosen based on the analysis goal and the characteristics of the design.Keywords
This publication has 13 references indexed in Scilit:
- Queueing Petri Nets-A formalism for the combined qualitative and quantitative analysis of systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Design of multimedia global PACS distributed computing environmentPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A probabilistic approach to parallel system performance modellingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- The model checker SPINIEEE Transactions on Software Engineering, 1997
- Estimating and optimizing performance for parallel programsComputer, 1995
- Understanding quality in conceptual modelingIEEE Software, 1994
- Predicting performance of parallel computationsIEEE Transactions on Parallel and Distributed Systems, 1990
- A modeling methodology for the analysis of concurrent systems and computationsJournal of Parallel and Distributed Computing, 1989
- Combining queueing networks and generalized stochastic Petri nets for the solution of complex models of system behaviorIEEE Transactions on Computers, 1988
- Performance and Reliability Analysis Using Directed Acyclic GraphsIEEE Transactions on Software Engineering, 1987