异构信号处理平台中层次性流水线调度算法
|
摘要
现有实时任务调度算法在系统异构性方面没有考虑处理节点计算能力的差异,导致任务划分不均衡。为此,根据异构信号处理平台实时任务的特点,提出一种面向异构信号处理平台的层次性流水线调度算法。基于数据流图的多层次任务划分算法并借鉴多层图划分思想,实现任务调度负载均衡和低通信同步开销。采用同步流水线调度方法达到低延迟实时数据处理的目的。实验结果表明,与事件触发数据流调度算法相比,该算法能有效提高异构信号处理平台的实时处理能力。
The existing real-time task scheduling algorithm do not consider the difference of node computing power in system heterogeneity,which leads to the imbalance of task partition. Therefore,according to the characteristics of realtime task in the heterogeneous signal processing platform,a multi-level pipeline scheduling algorithm for heterogeneous signal processing platform is proposed. The multi-level task partition algorithm based on data flow graph and the idea of multi-layer graph partition are used to realize the task scheduling load balancing and low communication synchronization overhead. The synchronous pipeline scheduling method is used to achieve the purpose of low delay real-time data processing. Experimental results show that,compared with the event-triggered data flow scheduling algorithm,this algorithm can effectively improve the real-time processing capability of heterogeneous signal processing platform.
The existing real-time task scheduling algorithm do not consider the difference of node computing power in system heterogeneity,which leads to the imbalance of task partition. Therefore,according to the characteristics of realtime task in the heterogeneous signal processing platform,a multi-level pipeline scheduling algorithm for heterogeneous signal processing platform is proposed. The multi-level task partition algorithm based on data flow graph and the idea of multi-layer graph partition are used to realize the task scheduling load balancing and low communication synchronization overhead. The synchronous pipeline scheduling method is used to achieve the purpose of low delay real-time data processing. Experimental results show that,compared with the event-triggered data flow scheduling algorithm,this algorithm can effectively improve the real-time processing capability of heterogeneous signal processing platform.
引文
[1]MITOLA J.The software radio architecture[J].IEEECommunications Magazine,1995,33(5):26-38.
[2]MAROJEVIC V,BALLESTX R,GELONCH A.Acomputing resource management framework for software-defined radios[J].IEEE Transactions on Computers,2008,57(10):1399-1412.
[3]ROY F L,LABABIDI R.A survey of frameworks and open environments applied to cognitive radio design[J].International Journal of Emerging Technology and Advanced Engineering,2014,4(3):356-364.
[4]TAN K,LIU H,ZHANG J.Sora:high-performance software radio using general-purpose multi-core processors[J].Communications of the ACM,2011,54(1):99-107.
[5]CHEN D,VANHOY G,BEAUFAIT M,et al.OSSIE/GNU radio generic component[C]//Proceedings of2011 Wireless Telecommunications Symposium.Washington D.C.,USA:IEEE Press,2011:1-5.
[6]MARLOW R L.Making radios with GReasy:GNUradio with FPGAs made easy[D].[S.l.]:Virginia Polytechnic Institute and State University,2014.
[7]李丹丹,马金全,杨平平.信号处理平台中可重构组件的设计与实现[J].计算机工程,2018,44(5):33-39.
[8]尹杨美,徐成,刘彦.改进的异构多处理器的实时任务调度算法研究[J].计算机应用研究,2010,27(4):1236-1238.
[9]MAROJEVIC V.Computing resource management in software-defined and cognitive radios[D].Barcelona,Spain:Universitat Politècnica de Catalunya,2009.
[10]GOMEZ I,MAROJEVIC V,GELONCH A.Aloe:an open-source SDR execution environment with cognitive computing resource management capabilities[J].IEEECommunications Magazine,2011,49(9):76-83.
[11]SRIRAM S,BHATTACHARYYA S S.Embedded multiprocessors:scheduling and synchronization[M].[S.l.]:CRC Press,2009.
[12]WEI H,YU J,YU H,et al.Software pipelining for stream programs on resource constrained multicore architectures[J].IEEE Transactions on Parallel and Distributed Systems,2012,23(12):2338-2350.
[13]WEI H,QIN M.StreamTMC:stream compilation for tiled multi-core architectures[J].Journal of Parallel and Distributed Computing,2013,73(4):484-494.
[14]于俊清,张维维,陈文斌,等.面向多核集群的数据流程序层次流水线并行优化方法[J].计算机学报,2014,37(10):2071-2083.
[15]沈轶炜,曾国荪.异构计算中一种图的非均衡划分算法[J].计算机科学,2006,33(6):260-263.
[16]MITOLA J.The software radio architecture[J].IEEECommunications Magazine,1995,33(5):26-38.
[17]贺靖卿,岳春生,胡泽明.考虑综合性约束的软件无线电应用部署算法[J].计算机工程与应用,2017,53(15):239-243.
[2]MAROJEVIC V,BALLESTX R,GELONCH A.Acomputing resource management framework for software-defined radios[J].IEEE Transactions on Computers,2008,57(10):1399-1412.
[3]ROY F L,LABABIDI R.A survey of frameworks and open environments applied to cognitive radio design[J].International Journal of Emerging Technology and Advanced Engineering,2014,4(3):356-364.
[4]TAN K,LIU H,ZHANG J.Sora:high-performance software radio using general-purpose multi-core processors[J].Communications of the ACM,2011,54(1):99-107.
[5]CHEN D,VANHOY G,BEAUFAIT M,et al.OSSIE/GNU radio generic component[C]//Proceedings of2011 Wireless Telecommunications Symposium.Washington D.C.,USA:IEEE Press,2011:1-5.
[6]MARLOW R L.Making radios with GReasy:GNUradio with FPGAs made easy[D].[S.l.]:Virginia Polytechnic Institute and State University,2014.
[7]李丹丹,马金全,杨平平.信号处理平台中可重构组件的设计与实现[J].计算机工程,2018,44(5):33-39.
[8]尹杨美,徐成,刘彦.改进的异构多处理器的实时任务调度算法研究[J].计算机应用研究,2010,27(4):1236-1238.
[9]MAROJEVIC V.Computing resource management in software-defined and cognitive radios[D].Barcelona,Spain:Universitat Politècnica de Catalunya,2009.
[10]GOMEZ I,MAROJEVIC V,GELONCH A.Aloe:an open-source SDR execution environment with cognitive computing resource management capabilities[J].IEEECommunications Magazine,2011,49(9):76-83.
[11]SRIRAM S,BHATTACHARYYA S S.Embedded multiprocessors:scheduling and synchronization[M].[S.l.]:CRC Press,2009.
[12]WEI H,YU J,YU H,et al.Software pipelining for stream programs on resource constrained multicore architectures[J].IEEE Transactions on Parallel and Distributed Systems,2012,23(12):2338-2350.
[13]WEI H,QIN M.StreamTMC:stream compilation for tiled multi-core architectures[J].Journal of Parallel and Distributed Computing,2013,73(4):484-494.
[14]于俊清,张维维,陈文斌,等.面向多核集群的数据流程序层次流水线并行优化方法[J].计算机学报,2014,37(10):2071-2083.
[15]沈轶炜,曾国荪.异构计算中一种图的非均衡划分算法[J].计算机科学,2006,33(6):260-263.
[16]MITOLA J.The software radio architecture[J].IEEECommunications Magazine,1995,33(5):26-38.
[17]贺靖卿,岳春生,胡泽明.考虑综合性约束的软件无线电应用部署算法[J].计算机工程与应用,2017,53(15):239-243.