Fuzzy multiprocessor architecture reconfiguration based on dynamic frequency scaling

摘要

Multicore systems are rapidly immerging as a new paradigm for integrating both high performance systems and embedded applications. A large field of research focuses now on reconfigurable multicore architecture in order to find optimal solutions for improved performances such as energy consumption. A Perfect system would be able to create a balance between optimum resources allocation and a particular workload. The dynamic frequency scaling has been a key technique in exploiting the hardware configurable characteristics of processors. However, for large class of applications in embedded real time systems, the variable operating frequency interferes with their deadline guarantee. In this paper, a novel fuzzy logic technique is presented to optimize real time system feasibility and energy consumption in multicore architecture based on operating frequency scaling. Tested on different periodic task sets generated with different system charges the proposed technique is an intelligent decision aid system that studies schedulability and energy consumption aware.