Simulation of Efficient Real-Time Scheduling and Power Optimisation

International audienceSophisticated applications turn out to be executed upon more than one CPU for practical and economic reasons. Due to advances in circuit technology and performance limitation, multi-core technology has become the mainstream in CPU designs. However, the most serious limitation of these devices is the battery lifetime since battery technology is not keeping up with the rest of the power-hungry processors and peripherals used in today's mobile devices. As a solution, many investigations have turned toward the algorithms of power management combined with some scheduling policies. They can make significant energy saving while preserving the temporal constraints of these embedded systems. Reducing energy, especially, affect not only the battery lifetime, but also aim to reduce the heat generated by real-time embedded controller in various products or even to decrease the conditions of cooling and the costs, in the large scale, of giant multiprocessor computers. To assess the behavior and performance of the strategy of scheduling a flexible multiprocessor scheduling simulation and evaluation platform is needed. This paper puts forth the claim that the STORM simulator improves application quality both in terms of execution time and energy consumption for a high performance mobile computing embedded system design