Programação >> Prof. A. Mounir Boukadoum ( University of Quebec at Montreal, Canadá) Speaker Prof. A. Mounir Boukadoum ( University of Quebec at Montreal, Canadá) Title Towards automated scheduling algorithm selection in MPSoC design tools Abstract The ongoing shrinkage of semiconductor geometries is allowing increasingly higher system-on-chip densities, with more and more embedded processors. As a result, task scheduling has become an important concern during the system design of embedded multi processor systems and research in this area has produced a substantial and diversified knowledge. The presentation will address two issues related to the efficient management and taping of this knowledge in the context of design automation: 1. the use of functional concept analysis to structure the available task scheduling information for optimal application to Multi Processor System on Chip (MPSoC) design; 2. The use of association rule learning to extract performance information about a given scheduling algorithm in the context of a set of applications. In both cases, the ultimate goal is to use knowledge and data mining techniques to automatically extract knowledge about scheduling algorithm, increase the designer's knowledge about their respective strengths and weaknesses, and allow the automatic selection of the scheduling algorithm best fit to a given application. SHORT CVs Mounir Boukadoum is Professor of Microelectronics Engineering at the department of computer science, Université du Québec à Montréal, Canada. After studying Physics at The University of Algiers (Algeria), he switched to electrical engineering and received the Master's degree from The Stevens Institute of Technology (USA) and the Ph.D. degree from The University of Houston (USA). Prof. Boukadoum has vested interest in many research areas, of which biomedical instrumentation and signal processing, artificial intelligence and design automation.