期刊文献+

高性能多媒体SoC分组访存调度算法 被引量:2

A Memory Group Scheduling Algorithm for High-performance Multimedia SoC
下载PDF
导出
摘要 根据多媒体处理单元的访存特点,提出一种面向高性能多媒体SoC的分组访存调度算法.该算法将访存请求按照访存ID和页地址分组,以访存组为单位进行乱序调度,并通过维护相同ID访存请求之间的顺序保证访存的正确性:综合考虑访存单元的访存效率和服务质量要求,在每个访存单元独立的调度周期内提供最低带宽保障服务.将该分组访存调度算法应用于访存调度装置,实际应用仿真结果表明,与已有基于带宽分配的访存调度算法相比,文中算法在保障访存单元带宽需求的同时降低了访存延迟,并将平均带宽利用率提高了15%. Based on the memory access characteristics of multimedia processing units,a memory group scheduling algorithm for high-performance multimedia SoC is proposed.The algorithm,whose scheduling unit is memory group,groups memory requests and reorders the memory groups in accordance with group ID and page address.To ensure the accuracy of memory access,memory requests with the same memory ID must be scheduled according to the order of arrival.The algorithm considers memory efficiency and quality-of-service together,and provides minimum bandwidth guarantee for every memory requester in each scheduling cycle.The group scheduling algorithm is applied to memory scheduling device and compared with the previous memory scheduling algorithms based on bandwidth allocation in the simulation environment under practical application.Experimental result shows that the proposed algorithm can provide better quality-of-service and achieve average bandwidth utilization improvement by 15%.
出处 《计算机辅助设计与图形学学报》 EI CSCD 北大核心 2011年第9期1585-1592,共8页 Journal of Computer-Aided Design & Computer Graphics
基金 国家"核高基"科技重大专项课题(2009ZX01028-002-003 2009ZX01029-001-003) 国家"八六三"高技术研究发展计划(2008AA010901) 国家自然科学基金(60736012 60921002 61070025)
关键词 访存调度 多媒体 片上系统 SDRAM memory scheduling multimedia system-on-chip SDRAM
  • 相关文献

参考文献13

  • 1Kollig P, Osborne C, Henriksson T. Heterogeneous multi-core platform for consumer muhimedia applications [C] //Proceedings of the Con{erence on Design, Automation and Test in Europe. New York: ACMPress, 2009:1254-1259.
  • 2Lin T C, Lee K B, Jen C W. Quality-aware memory controller for multimedia platform SoC [C] //Proceedings of IEEE Workshop on Signal Processing Systems. Los Alamitos: IEEE Computer Society Press, 2003:328-333.
  • 3Lee K B, Lin T C, Jen C W. An efficient quality aware memory controller for multimedia platform SoC [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2005, 15(5): 620-633.
  • 4Rixner S, Dally W J, Kapasi U J, et al. Memory access scheduling [C] //Proceedings of the 27th International Symposium on Computer Architecture, Los Alamitos: IEEE Computer Society Press, 2000:128-138.
  • 5Shao J, Davis B T. A burst scheduling access reordering mechanism[C] //Proceedings of IEEE lath International Symposium on High Performance Computer Architecture. Los Alamitos: IEEE Computer Society Press, 2007:285-294.
  • 6Pang J, Yang L, Shi L, et al, A priority-expression -based burst scheduling of memory reordering access [C] //Proceedings of International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation. Los Alamitos: IEEE Computer Society Press, 2008:203-209.
  • 7Shi L, Pang J, Yang L, etal. Fair priority expression-based burst scheduling to enhance performance and fairness of shared dram systems [C]//Proceedings of Canadian Conference on Electrical and Computer Engineering. Los Alamitos: IEEE Computer Society Press, 2009:190-194.
  • 8Su M H, Gao X, Chen Y J, et al. Efficiency-aware QoS DRAM scheduler [C] //Proceedings of IEEE International Conference on Networking, Architecture, and Storage. Los Alamitos:IEEE Computer Society Press, 2009:223-228.
  • 9Zhu J Y, Liu P L, Zhou D J. An SDRAM controller optimized for high definition video coding application[C] //Proceedings of IEEE International Symposium on Circuits and Systems. Los Alamitos: IEEE Computer Society Press, 2008:3518-3521.
  • 10der Wolf P, Henriksson T. Video processing requirements on SoC infrastructures [C] //Proceedings of Design, Automation and Test in Europe. Los Alamitos: IEEE Computer Society Press, 2008:1124-1125.

同被引文献40

  • 1李中,王刚,刘璟.基于服务质量需求的存储系统动态优先权调度策略[J].计算机研究与发展,2006,43(11):2027-2032. 被引量:1
  • 2Lee C 1, Narasiman V, Mutlu 0, et al. Improving memory bank-level parallelism in the presence of prefetching[C] II Proceedings of the 42nd Annual IEEEI ACM International Symposium on Microarchitecture. New York: ACM Press, 2009: 327-336.
  • 3Lin C H, Yang C L, King K 1. PPT: joint performancel powerlthermal management of DRAM memory for multi-core systems[C] II Proceedings of the 14th ACMlIEEE International Symposium on Low Power Electronics and Design. New York: ACM Press, 2009: 93-98.
  • 4Amit N, Ben-Yehuda M, Tsafrir D, et al. vIOMMU: efficient IOMMU emulation[C] II Proceedings of USE NIX Conference on USENIX Annual Technical Conference. Berkeley: USENIX Press, 2011: 73-86.
  • 5Budnik P, Kuck D J. The organization and use of parallel memories[J]. IEEE Transactions on Computers, 1971,20(12): 1566-1569.
  • 6Lawrie D H, Vora C R. The prime memory system for array access[I]. IEEE Transactions on Computers, 1982, 31(5): 435-442.
  • 7Sudan K, Chatterjee N, Nellans D, et al. Micro-pages: increasing DRAM efficiency with locality-aware data place?ment[C] II Proceedings of the 15th International Conference on Architectural Support for Programming Languages and Oper?ating Systems. New York: ACM Press, 2010: 219-230.
  • 8Yu G S, Chang T S. Optimal data mapping for motion compensation in H.264 video decoding[C] II Proceeding of the IEEE Workshop on Signal Processing Systems. Los Alamitos: IEEE Computer Society Press, 2007: 505-508.
  • 9Zhu 1 Y, Liu P L, Zhou D 1. An SDRAM controller optimized for high definition video coding application[C] II Proceeding ofIEEE International Symposium on Circuits and Systems. Los Alamitos: IEEE Computer Society Press, 2008: 3518-3521.
  • 10Jeong M K, Yoon D H, Sunwoo D, et al. Balancing DRAM locality and parallelism in shared memory CMP systems[C] II Proceedings of the 18th IEEE International Symposium on High-Performance Computer Architecture. Los Alamitos: IEEE Computer Society Press, 2012: 1-12.

引证文献2

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部