异构多处理器SoC的应用算法性能优化方法  

Performance Optimization of Application Algorithms for Heterogeneous Multi-Processor System-on-Chips

作　　者：赵鹏[1] 严明[1] 李思昆[1] 

机构地区：[1]国防科学技术大学计算机学院,湖南长沙410073

出　　处：《软件学报》2011年第7期1475-1487,共13页Journal of Software

基　　金：国家自然科学基金(90207019;90707003)

摘　　要：在嵌入式多媒体处理领域中,多处理器片上系统(multi-processor system-on-chip,简称MPSoC)的应用越来越广泛.多媒体处理MPSoC通常采用"主处理器核+多个异构协处理器核"的主流体系结构.该结构兼顾了MPSoC系统的通用性与灵活性、性能与功耗,但也向MPSoC的性能优化方法提出了更高的要求.针对异构MPSoC上的多媒体应用算法,提出了一种MPSoC多媒体处理性能优化方法.该方法经过应用特征分析、循环仿射划分、应用向MPSoC各处理器核的映射,实现了优化的数据局部性与多级并行性,从而提高了异构MPSoC上多媒体应用算法的性能.实验结果表明,该方法对于多媒体应用算法在异构MPSoC上的处理性能优化方面取得了明显效果.MPSoCs （multi-processor system-on-chips） are comprehensively applied in the embedded multimediaprocessing field. Multimedia MPSoCs often adopt the ＂host processor＋multiple heterogeneous synergistic processor＂ architecture, which makes the trade-off between universality and flexibility. MPSoCs also take into consideration both performance and power-consumption, but challenges the method of performance optimization of System-on-Chip applications. This paper proposes an approach that improves the performance of application algorithms running on heterogeneous MPSoCs. The approach includes three stages： Application feature analysis, affine partitioning of kernel loops, and ＂application-architecture＂ mapping. It optimizes the multi-level parallelism and data locality of application algorithms to improve MPSoC performance. Experimental results show that the proposed approach can greatly improve the multimedia processing performance on heterogeneous MPSoCs.

关 键 词：片上系统 多处理器片上系统 嵌入式系统 多媒体处理 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]