基于框架的虚拟计算环境关键技术研究
|
摘要
通过分析目前几种虚拟计算环境的现状、特点及适用范围,针对中小规模、异构零散的的计算需求,提出一种基于框架技术的虚拟计算环境的实现方法,并结合框架技术设计了分层服务模型.基于复杂多样的、性能差距明显的计算资源的调度问题,提出了一种基于对等模式的随机自适应调度算法.该算法提供对关联任务的调度,适合异构的、资源规模较大的环境,尤其是计算资源差异明显的异构计算环境.实现了基于框架技术的虚拟计算原型系统,并通过实验测试对比算法的调度性能,最后将计算环境应用于实际的地震资料处理中.
By analyzing the current situation of serveral virtual environments,characteristics and scope of application,This paper designs and implements a framework-based virtual computing environment for small and medium-sized,heterogeneous fragmented computing needs.presents a hierarchical service design to improve service scalability and flexibility,and proposes a peer-to-peer mode random adaptive scheduling algorithm for the diverse complicated performance gap between computing resources scheduling problems.This algorithm supports to scheduler related task.the algorithm is applicable to heterogeneous and larger resource environment,particularly on the computing resources performance difference.The paper builds a prototype system and implement experiment,Finally,the computing environment is used in the actual seismic data processing.
By analyzing the current situation of serveral virtual environments,characteristics and scope of application,This paper designs and implements a framework-based virtual computing environment for small and medium-sized,heterogeneous fragmented computing needs.presents a hierarchical service design to improve service scalability and flexibility,and proposes a peer-to-peer mode random adaptive scheduling algorithm for the diverse complicated performance gap between computing resources scheduling problems.This algorithm supports to scheduler related task.the algorithm is applicable to heterogeneous and larger resource environment,particularly on the computing resources performance difference.The paper builds a prototype system and implement experiment,Finally,the computing environment is used in the actual seismic data processing.
引文
[1]Lu X,Wang H.Internet-based virtual computing environment:con-cepts and architecture[J].Information Sciences,2006:49(6)681-701.
[2]Youseff L.Toward a unified ontology of cloud computing[C].GridComputing Environments Workshop,2008,GCE,2008:1-10.
[3]Foster I.Cloud computing and grid computing 360-degree compared[C].Grid Computing Environments Workshop,GCE,2008:12-16.
[4]Peter Mell,Timothy Grance.The NIST definition of cloud compu-ting[S].National Institute of Standards and Technology SpecialPublication(800-145),2011.
[5]Daniel Nurmi,Rich Wolski,Chris Grzegorczyk,et al.The eucalyp-tus open-source cloud-computing system[C].9th IEEE/ACM In-ternational Symposium on Cluster Computing and the Grid,2009:124-132.
[6]Rod Johnson,Juergen Hoeller,Alef Arendsen,et al.Professional ja-va development w ith the spring framew ork[M].Wrox Press,2005.
[7]Rod Johnson,Juergen hoeller,Keith Donald,et al.Spring frame-w ork reference document[EB/OL].http://w w w.springsource.org,2011.
[8]Jiang Ting-yao,Li Qing-hua.A scheduling algorithm for DAG taskgraphs[J].Mini-Micro Systems,2003,24(10):1796-1799.
[9]Qi Xue-dong,Li Xia,Liang Hong.Implement of grid compled par-allel programming strategy in seismic forw ard modeling[J].Journalof PLA University of Science and Technology,2008,9(5):502-506.
[10]Ellert M,Gronager M,Konstantinov A,et al.Advanced resourceconnector middlew are for lightw eight computational grids[J].Fu-ture Generation Computer Systems,2007,23(2):219-240.
[11]Li Jian-jiang,Shu Ji-wu,Wang You-xin.A parellel algorithm forprestack depth migratim based on shared memory[J].Journal ofSoftw are,2002,13(12):2231-2236.
[12]Zhang Shu-fen,Zhang Shuai,Chen Xue-bin,et al.Analysis and re-search of cloud computing system instance[C].Second Interna-tional Conference on Future Netw orks,2010:88-92.
[8]蒋廷耀,李庆华.DAG任务图的一种调度算法[J].小型微型计算机系统,2003,24(10):1796-1799.
[9]亓雪东,李霞,梁鸿.网格并行编程方法及其在地震波场正演模拟中的应用[J].解放军理工大学学报(自然科学版),2008,9(5):502-506.
[11]李建江,舒继武,王有新,等.一种基于共享存储的叠前深度偏移并行算法[J].软件学报,2002,13(12):2231-2236.
[2]Youseff L.Toward a unified ontology of cloud computing[C].GridComputing Environments Workshop,2008,GCE,2008:1-10.
[3]Foster I.Cloud computing and grid computing 360-degree compared[C].Grid Computing Environments Workshop,GCE,2008:12-16.
[4]Peter Mell,Timothy Grance.The NIST definition of cloud compu-ting[S].National Institute of Standards and Technology SpecialPublication(800-145),2011.
[5]Daniel Nurmi,Rich Wolski,Chris Grzegorczyk,et al.The eucalyp-tus open-source cloud-computing system[C].9th IEEE/ACM In-ternational Symposium on Cluster Computing and the Grid,2009:124-132.
[6]Rod Johnson,Juergen Hoeller,Alef Arendsen,et al.Professional ja-va development w ith the spring framew ork[M].Wrox Press,2005.
[7]Rod Johnson,Juergen hoeller,Keith Donald,et al.Spring frame-w ork reference document[EB/OL].http://w w w.springsource.org,2011.
[8]Jiang Ting-yao,Li Qing-hua.A scheduling algorithm for DAG taskgraphs[J].Mini-Micro Systems,2003,24(10):1796-1799.
[9]Qi Xue-dong,Li Xia,Liang Hong.Implement of grid compled par-allel programming strategy in seismic forw ard modeling[J].Journalof PLA University of Science and Technology,2008,9(5):502-506.
[10]Ellert M,Gronager M,Konstantinov A,et al.Advanced resourceconnector middlew are for lightw eight computational grids[J].Fu-ture Generation Computer Systems,2007,23(2):219-240.
[11]Li Jian-jiang,Shu Ji-wu,Wang You-xin.A parellel algorithm forprestack depth migratim based on shared memory[J].Journal ofSoftw are,2002,13(12):2231-2236.
[12]Zhang Shu-fen,Zhang Shuai,Chen Xue-bin,et al.Analysis and re-search of cloud computing system instance[C].Second Interna-tional Conference on Future Netw orks,2010:88-92.
[8]蒋廷耀,李庆华.DAG任务图的一种调度算法[J].小型微型计算机系统,2003,24(10):1796-1799.
[9]亓雪东,李霞,梁鸿.网格并行编程方法及其在地震波场正演模拟中的应用[J].解放军理工大学学报(自然科学版),2008,9(5):502-506.
[11]李建江,舒继武,王有新,等.一种基于共享存储的叠前深度偏移并行算法[J].软件学报,2002,13(12):2231-2236.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心