DeSi: A Visual Tool for Analyzing Distributed Software Architectures

Poster

Co-Researcher: Marija Mikic-Rakic, Sam Malek

Abstract: The deployment architecture of a distributed system can have a direct impact on certain non-functional properties of that system. The non-functional properties depend on various system parameters such as network bandwidth, network reliability, the frequency of component interactions and so on. Currently, no tool exists for assessing the impact of these system parameters on distributed system properties such as availability or latency. In this poster, we will present DeSi, a visual tool that allows us to specify system parameters and deployments and then analyze their effect on the overall system availability. At the same time, DeSi can also interface with an architectural middleware, allowing us to monitor a running system and to effect component redeployment on that running system.

Bio: Nels Beckman is an undergraduate student at the University of Southern California. He is currently in his fourth year and plans to pursue his interests in distributed systems and software architecture as a doctoral student after the completion of his undergraduate studies.