为了更好地满足云计算中用户的服务质量(quality of service,QoS)需求,合理利用云数据中心的资源,以任务的执行时间和虚拟机的负载均衡作为优化的目标对象,提出了一种基于烟花算法(fireworks algorithm,FWA)的多目标优化调度模型。烟花...为了更好地满足云计算中用户的服务质量(quality of service,QoS)需求,合理利用云数据中心的资源,以任务的执行时间和虚拟机的负载均衡作为优化的目标对象,提出了一种基于烟花算法(fireworks algorithm,FWA)的多目标优化调度模型。烟花算法是一种启发式算法,利用爆炸算子、高斯变异和选择策略能较快地寻找到全局最优解。通过在Cloudsim上与粒子群优化算法(PSO)和遗传算法(GA)进行有效性和执行时间上的对比,结果表明烟花算法在不同实验次数下可持续得到最优适应度值,而且在种群规模不断扩大时,烟花算法的执行时间没有陡然增加,明显优于PSO算法和GA算法。展开更多
针对目前时间敏感网络(Time Sensitive Network,TSN)中多采用离线调度,在处理动态需求时调度开销过大的问题,提出了一种适用于TSN的基于整数线性规划(Integer Linear Programming,ILP)的动态流量均衡调度算法.该算法以已生成的离线调度...针对目前时间敏感网络(Time Sensitive Network,TSN)中多采用离线调度,在处理动态需求时调度开销过大的问题,提出了一种适用于TSN的基于整数线性规划(Integer Linear Programming,ILP)的动态流量均衡调度算法.该算法以已生成的离线调度表为基础,通过一种增量计算方式,有效降低拓扑和业务变化带来的开销;同时,针对链路负载可能出现的极端情况,制订了流量均衡策略,避免出现延时瓶颈问题,进而提升通信实时性.实验数据表明,与传统静态调度算法相比,在处理动态流量时的运算时间减少,链路负载更加均衡.展开更多
Parallel computing techniques have been introduced into digital image correlation(DIC) in recent years and leads to a surge in computation speed. The graphics processing unit(GPU)-based parallel computing demonstrated...Parallel computing techniques have been introduced into digital image correlation(DIC) in recent years and leads to a surge in computation speed. The graphics processing unit(GPU)-based parallel computing demonstrated a surprising effect on accelerating the iterative subpixel DIC, compared with CPU-based parallel computing. In this paper, the performances of the two kinds of parallel computing techniques are compared for the previously proposed path-independent DIC method, in which the initial guess for the inverse compositional Gauss-Newton(IC-GN) algorithm at each point of interest(POI) is estimated through the fast Fourier transform-based cross-correlation(FFT-CC) algorithm. Based on the performance evaluation, a heterogeneous parallel computing(HPC) model is proposed with hybrid mode of parallelisms in order to combine the computing power of GPU and multicore CPU. A scheme of trial computation test is developed to optimize the configuration of the HPC model on a specific computer. The proposed HPC model shows excellent performance on a middle-end desktop computer for real-time subpixel DIC with high resolution of more than 10000 POIs per frame.展开更多
作为流式大数据计算的主要平台之一,Storm在设计过程中由于缺乏节能的考虑,导致其存在高能耗与低效率的问题.传统的节能策略并未考虑Storm的性能约束,可能会对集群的实时性造成影响.针对这一问题,设计了资源约束模型、最优线程重分配模...作为流式大数据计算的主要平台之一,Storm在设计过程中由于缺乏节能的考虑,导致其存在高能耗与低效率的问题.传统的节能策略并未考虑Storm的性能约束,可能会对集群的实时性造成影响.针对这一问题,设计了资源约束模型、最优线程重分配模型以及数据迁移模型.进一步提出了Storm平台下的线程重分配与数据迁移节能策略(energy-efficient strategy based on executor reallocation and data migration in Storm,简称ERDM),包括资源约束算法与数据迁移算法.其中,资源约束算法根据集群各工作节点CPU、内存与网络带宽的资源占用率,判断集群是否允许数据的迁移.数据迁移算法根据资源约束模型与最优线程重分配模型,设计了数据迁移的最优化方法.此外,ERDM通过分配线程减少了节点间的通信开销,并根据大数据流式计算的性能与能效评估ERDM.实验结果表明,与现有研究相比,ERDM能够有效降低节点间通信开销与能耗,并提高集群的性能.展开更多
文摘为了更好地满足云计算中用户的服务质量(quality of service,QoS)需求,合理利用云数据中心的资源,以任务的执行时间和虚拟机的负载均衡作为优化的目标对象,提出了一种基于烟花算法(fireworks algorithm,FWA)的多目标优化调度模型。烟花算法是一种启发式算法,利用爆炸算子、高斯变异和选择策略能较快地寻找到全局最优解。通过在Cloudsim上与粒子群优化算法(PSO)和遗传算法(GA)进行有效性和执行时间上的对比,结果表明烟花算法在不同实验次数下可持续得到最优适应度值,而且在种群规模不断扩大时,烟花算法的执行时间没有陡然增加,明显优于PSO算法和GA算法。
文摘针对目前时间敏感网络(Time Sensitive Network,TSN)中多采用离线调度,在处理动态需求时调度开销过大的问题,提出了一种适用于TSN的基于整数线性规划(Integer Linear Programming,ILP)的动态流量均衡调度算法.该算法以已生成的离线调度表为基础,通过一种增量计算方式,有效降低拓扑和业务变化带来的开销;同时,针对链路负载可能出现的极端情况,制订了流量均衡策略,避免出现延时瓶颈问题,进而提升通信实时性.实验数据表明,与传统静态调度算法相比,在处理动态流量时的运算时间减少,链路负载更加均衡.
基金supported by the National Natural Science Foundation of China(Grant Nos.11772131,11772132,11772134&11472109)the Natural Science Foundation of Guangdong Province,China(Grant Nos.2015A030308017,2015A030311046&2015B010131009)+2 种基金the Opening fund of State Key Laboratory of Nonlinear Mechanics(LNM)CASthe State Key Lab of Subtropical Building Science,South China University of Technology(Grant Nos.2014ZC17&2017ZD096)
文摘Parallel computing techniques have been introduced into digital image correlation(DIC) in recent years and leads to a surge in computation speed. The graphics processing unit(GPU)-based parallel computing demonstrated a surprising effect on accelerating the iterative subpixel DIC, compared with CPU-based parallel computing. In this paper, the performances of the two kinds of parallel computing techniques are compared for the previously proposed path-independent DIC method, in which the initial guess for the inverse compositional Gauss-Newton(IC-GN) algorithm at each point of interest(POI) is estimated through the fast Fourier transform-based cross-correlation(FFT-CC) algorithm. Based on the performance evaluation, a heterogeneous parallel computing(HPC) model is proposed with hybrid mode of parallelisms in order to combine the computing power of GPU and multicore CPU. A scheme of trial computation test is developed to optimize the configuration of the HPC model on a specific computer. The proposed HPC model shows excellent performance on a middle-end desktop computer for real-time subpixel DIC with high resolution of more than 10000 POIs per frame.
文摘作为流式大数据计算的主要平台之一,Storm在设计过程中由于缺乏节能的考虑,导致其存在高能耗与低效率的问题.传统的节能策略并未考虑Storm的性能约束,可能会对集群的实时性造成影响.针对这一问题,设计了资源约束模型、最优线程重分配模型以及数据迁移模型.进一步提出了Storm平台下的线程重分配与数据迁移节能策略(energy-efficient strategy based on executor reallocation and data migration in Storm,简称ERDM),包括资源约束算法与数据迁移算法.其中,资源约束算法根据集群各工作节点CPU、内存与网络带宽的资源占用率,判断集群是否允许数据的迁移.数据迁移算法根据资源约束模型与最优线程重分配模型,设计了数据迁移的最优化方法.此外,ERDM通过分配线程减少了节点间的通信开销,并根据大数据流式计算的性能与能效评估ERDM.实验结果表明,与现有研究相比,ERDM能够有效降低节点间通信开销与能耗,并提高集群的性能.
