基于OpenFlow的SDN(Software Defined Networking)技术在数据中心中得到广泛研究和应用,如何缓解集中的控制平面成为网络性能的瓶颈是其中的研究热点.OpenFlow规范提出,当数据平面有缓存能力时,未命中的报文仅需发送少量摘要信息至控制...基于OpenFlow的SDN(Software Defined Networking)技术在数据中心中得到广泛研究和应用,如何缓解集中的控制平面成为网络性能的瓶颈是其中的研究热点.OpenFlow规范提出,当数据平面有缓存能力时,未命中的报文仅需发送少量摘要信息至控制器触发规则下发,从而减少控制平面与数据平面的通信负载.然而,现有的缓存模型采用报文粒度的缓存方式,使得同一条流的多个未命中报文会被送至控制器造成额外的通信负载,而且交换机处理报文的顺序会导致流内报文乱序,从而降低通信的性能.针对上述问题,该文提出了一种支持流内报文保序的OpenFlow交换机流缓存管理模型.通过基于流粒度的未命中报文缓存方式,进一步减少控制平面与数据平面的通信开销.通过设计流动作预处理机制,实现同一条流内报文传输保序.该文分别基于软件交换机OFSoftSwitch与硬件网络实验平台NetMagic对该流缓存管理模型进行了原型系统验证.展开更多
An effective discrete artificial bee colony(DABC) algorithm is proposed for the flow shop scheduling problem with intermediate buffers(IBFSP) in order to minimize the maximum completion time(i.e makespan). The effecti...An effective discrete artificial bee colony(DABC) algorithm is proposed for the flow shop scheduling problem with intermediate buffers(IBFSP) in order to minimize the maximum completion time(i.e makespan). The effective combination of the insertion and swap operator is applied to producing neighborhood individual at the employed bee phase. The tournament selection is adopted to avoid falling into local optima, while, the optimized insert operator embeds in onlooker bee phase for further searching the neighborhood solution to enhance the local search ability of algorithm. The tournament selection with size 2 is again applied and a better selected solution will be performed destruction and construction of iterated greedy(IG) algorithm, and then the result replaces the worse one. Simulation results show that our algorithm has a better performance compared with the HDDE and CHS which were proposed recently. It provides the better known solutions for the makespan criterion to flow shop scheduling problem with limited buffers for the Car benchmark by Carlier and Rec benchmark by Reeves. The convergence curves show that the algorithm not only has faster convergence speed but also has better convergence value.展开更多
在轴向柱塞泵的工作过程中,通过配流盘进行高低压的循环切换,而配流盘的过流面积是影响柱塞泵内部压力平稳过渡的重要参数。为了完成对配流结构过流面积的求解计算,提出了一种基于多维流场点云的过流面积计算方法。首先,对配流盘流场的...在轴向柱塞泵的工作过程中,通过配流盘进行高低压的循环切换,而配流盘的过流面积是影响柱塞泵内部压力平稳过渡的重要参数。为了完成对配流结构过流面积的求解计算,提出了一种基于多维流场点云的过流面积计算方法。首先,对配流盘流场的点云化过程进行了理论分析,得到了点云化处理的实施方法;然后,对配流盘三维流域模型进行了系列修正,提出了缓冲槽连通阶段模型的修正方法,提高了过流面积计算的准确性;最后,对不同配流阶段的过流面积进行了计算,并与理论测量所得结果以及采用计算流体力学(CFD)数值模拟方法所得的结果,分别进行了对比。计算结果表明:基于多维流场点云的过流面积识别算法求得的过流面积峰值大小为174.5 mm 2,而理论测量值的大小为177.5 mm 2;二者的相对误差率大约为1.7%;而与基于CFD方法所得结果的整体相对误差率,大约在5%以内。研究结果表明:采用该方法,能够直接且较精准地完成对配流盘过流面积的求解与计算,有效地简化了过流面积变换趋势的求解过程。展开更多
文摘基于OpenFlow的SDN(Software Defined Networking)技术在数据中心中得到广泛研究和应用,如何缓解集中的控制平面成为网络性能的瓶颈是其中的研究热点.OpenFlow规范提出,当数据平面有缓存能力时,未命中的报文仅需发送少量摘要信息至控制器触发规则下发,从而减少控制平面与数据平面的通信负载.然而,现有的缓存模型采用报文粒度的缓存方式,使得同一条流的多个未命中报文会被送至控制器造成额外的通信负载,而且交换机处理报文的顺序会导致流内报文乱序,从而降低通信的性能.针对上述问题,该文提出了一种支持流内报文保序的OpenFlow交换机流缓存管理模型.通过基于流粒度的未命中报文缓存方式,进一步减少控制平面与数据平面的通信开销.通过设计流动作预处理机制,实现同一条流内报文传输保序.该文分别基于软件交换机OFSoftSwitch与硬件网络实验平台NetMagic对该流缓存管理模型进行了原型系统验证.
基金Projects(61174040,61104178,61374136) supported by the National Natural Science Foundation of ChinaProject(12JC1403400) supported by Shanghai Commission of Science and Technology,ChinaProject supported by the Fundamental Research Funds for the Central Universities,China
文摘An effective discrete artificial bee colony(DABC) algorithm is proposed for the flow shop scheduling problem with intermediate buffers(IBFSP) in order to minimize the maximum completion time(i.e makespan). The effective combination of the insertion and swap operator is applied to producing neighborhood individual at the employed bee phase. The tournament selection is adopted to avoid falling into local optima, while, the optimized insert operator embeds in onlooker bee phase for further searching the neighborhood solution to enhance the local search ability of algorithm. The tournament selection with size 2 is again applied and a better selected solution will be performed destruction and construction of iterated greedy(IG) algorithm, and then the result replaces the worse one. Simulation results show that our algorithm has a better performance compared with the HDDE and CHS which were proposed recently. It provides the better known solutions for the makespan criterion to flow shop scheduling problem with limited buffers for the Car benchmark by Carlier and Rec benchmark by Reeves. The convergence curves show that the algorithm not only has faster convergence speed but also has better convergence value.
文摘在轴向柱塞泵的工作过程中,通过配流盘进行高低压的循环切换,而配流盘的过流面积是影响柱塞泵内部压力平稳过渡的重要参数。为了完成对配流结构过流面积的求解计算,提出了一种基于多维流场点云的过流面积计算方法。首先,对配流盘流场的点云化过程进行了理论分析,得到了点云化处理的实施方法;然后,对配流盘三维流域模型进行了系列修正,提出了缓冲槽连通阶段模型的修正方法,提高了过流面积计算的准确性;最后,对不同配流阶段的过流面积进行了计算,并与理论测量所得结果以及采用计算流体力学(CFD)数值模拟方法所得的结果,分别进行了对比。计算结果表明:基于多维流场点云的过流面积识别算法求得的过流面积峰值大小为174.5 mm 2,而理论测量值的大小为177.5 mm 2;二者的相对误差率大约为1.7%;而与基于CFD方法所得结果的整体相对误差率,大约在5%以内。研究结果表明:采用该方法,能够直接且较精准地完成对配流盘过流面积的求解与计算,有效地简化了过流面积变换趋势的求解过程。
基金Supported by the National High-Tech Research and Development Plan of China under Grant No.2005AA121210(国家高技术研究发展计划(863))the National Basic Research Program of China under Grant No.2007CB307102(国家重点基础研究发展计划(973))
文摘针对流竞争拥塞,提出了一种拥塞分析模型FCCM(flow-competing congestion model),给出了TCP竞争流在拥塞链路上的分布特性,推导了流竞争拥塞发生的条件,进而分析了在流竞争拥塞发生时,路由器为维持拥塞链路100%利用率所需的最小缓存.分析结果表明,当流数目不确定时,应对流竞争拥塞所需的缓存将不大于流数目确定时经典BSCL(buffer sizing for congested interact links)方案中的最小缓存需求.
