中文摘要：

研究片上多核处理器系统的性能．功耗问题有两个重要因素：任务的执行时间以及系统的能量消耗．通过对CMP系统任务调度和能量消耗的分析建立了新颖的编码策略，并使用随机权重适应度以及精华解保留策略对粒子群优化算法进行改进，提出了多目标粒子群算法（MPSO）．仿真实验结果表明使用MPSO算法可以增加CMP系统中任务调度的效率，降低任务运行时间和系统能耗．

英文摘要：

The power-performance issues of chip multi-pmcessor（CMP） system have two important factors： the execution time of tasks and the system energy consumption. According to the analysis of CMP system tasks scheduling and energy consumption, a novel coding scheme is proposed. The particle swarm optimization （PSO） is improved by using the elite preserving strategy and the random weight fitness.And a multi-objective particle swarm optimization （MIPSO） is presented. Simulation results demonstrate that using MIPSO can increase the efficiency of task scheduling on CMP,decrease the execution time and energy consumption of system.