随着大数据应用的不断丰富,现在的数据中心通常部署着多种集群计算框架,并由统一的集群资源管理器(如Mesos)进行管理.目前的集群资源管理主要关注计算资源和存储资源,较少的涉及网络资源.但研究表明高效的网络资源管理对于优化作业性能...随着大数据应用的不断丰富,现在的数据中心通常部署着多种集群计算框架,并由统一的集群资源管理器(如Mesos)进行管理.目前的集群资源管理主要关注计算资源和存储资源,较少的涉及网络资源.但研究表明高效的网络资源管理对于优化作业性能十分重要.本文提出了一种基于SDN(Software Defined Network)的数据中心网络资源调度机制,该机制可以根据管理员预设的网络资源分配策略,加权的进行网络资源调度,为高优先级的作业分配更多网络资源以优化性能,并且实现不同作业之间的网络性能隔离.我们基于开源SDN控制器实现了原型系统,并通过实验验证了该机制的有效性.展开更多
The network resource allocation in SDN for control applications is becoming a key problem in the near future because of the conflict between the need of the flow-level flexibility control and the limited capacity of f...The network resource allocation in SDN for control applications is becoming a key problem in the near future because of the conflict between the need of the flow-level flexibility control and the limited capacity of flow table.Based on the analysis of the difference of the definition of network resource between SDN and traditional IP network,the idea of the integrated allocation of link bandwidth and flow table for multiple control applications in SDN is proposed in this paper.Furthermore,a price-based joint allocation model of network resource in SDN is built by introducing the price for each of the resources,which can get the proportional fair allocation of link bandwidth and the minimum global delay at the same time.We have also designed a popular flow scheduling policy based on the proportional fair allocation of link bandwidth in order to achieve the minimum global delay.A flow scheduling module has been implemented and evaluated in Floodlight,named virtual forwarding space(VFS).VFS can not only implement the fair allocation of link bandwidth and minimum delay flow scheduling in data plane but also accelerate packet forwarding by looking up control rules in control plane.展开更多
In thsssse cellular network, Relay Stations (RSs) help to improve the system performance; however, little work has been done considering the fairness of RSs. In this paper, we study the cooperative game approaches for...In thsssse cellular network, Relay Stations (RSs) help to improve the system performance; however, little work has been done considering the fairness of RSs. In this paper, we study the cooperative game approaches for scheduling in the wireless relay networks with two-virtual-antenna array mode. After defining the metric of relay channel capacity, we form a cooperative game for scheduling and present the interpretation of three different utilization objectives physically and mathematically. Then, a Nash Bargaining Solution (NBS) is utilized for resource allocation considering the traffic load fairness for relays. After proving the existence and uniqueness of NBS in Cooperative Game (CG-NBS), we are able to resolve the resource allocation problem in the cellular relay network by the relay selection and subcarrier assignment policy and the power allocation algorithm for both RSs and UEs. Simulation results reveal that the proposed CG-NBS scheme achieves better tradeoff between relay fairness and system throughput than the conventional Maximal Rate Optimization and Maximal Minimal Fairness methods.展开更多
Resource allocation is an important problem in ubiquitous network. Most of the existing resource allocation methods considering only wireless networks are not suitable for the ubiquitous network environment, and they ...Resource allocation is an important problem in ubiquitous network. Most of the existing resource allocation methods considering only wireless networks are not suitable for the ubiquitous network environment, and they will harm the interest of individual users with instable resource requirements. This paper considers the multi-point video surveillance scenarios in a complex network environment with both wired and wireless networks. We introduce the utility estimated by the total costs of an individual network user. The problem is studied through mathematical modeling and we propose an improved problem-specific branch-and-cut algorithm to solve it. The algorithm follows the divide-and-conquer principle and fully considers the duality feature of network selection. The experiment is conducted by simulation through C and Lingo. And it shows that compared with a centralized random allocation scheme and a cost greed allocation scheme, the proposed scheme has better per- formance of reducing the total costs by 13.0% and 30.6% respectively for the user.展开更多
Resource scheduling algorithm for ForCES(Forwarding and Control Element Separation) networks need to meet the flexibility,programmability and scalability of node resources.DBC(Deadline Budget Constrain) algorithm reli...Resource scheduling algorithm for ForCES(Forwarding and Control Element Separation) networks need to meet the flexibility,programmability and scalability of node resources.DBC(Deadline Budget Constrain) algorithm relies on users select cost or time priority,then scheduling to meet the requirements of users.However,this priority strategy of users is relatively simple,and cannot adapt to dynamic change of resources,it is inevitable to reduce the QoS.In order to improve QoS,we refer to the economic model and resource scheduling model of cloud computing,use SAL(Service Level Agreement) as pricing strategy,on the basis of DBC algorithm,propose an DABP(Deadline And Budget Priority based on DBC) algorithm for ForCES networks,DABP combines both budget and time priority to scheduling.In simulation and test,we compare the task finish time and cost of DABP algorithm with DP(Deadline Priority) algorithm and BP(Budget Priority) algorithm,the analysis results show that DABP algorithm make the task complete with less cost within deadline,benifical to load balancing of ForCES networks.展开更多
文摘随着大数据应用的不断丰富,现在的数据中心通常部署着多种集群计算框架,并由统一的集群资源管理器(如Mesos)进行管理.目前的集群资源管理主要关注计算资源和存储资源,较少的涉及网络资源.但研究表明高效的网络资源管理对于优化作业性能十分重要.本文提出了一种基于SDN(Software Defined Network)的数据中心网络资源调度机制,该机制可以根据管理员预设的网络资源分配策略,加权的进行网络资源调度,为高优先级的作业分配更多网络资源以优化性能,并且实现不同作业之间的网络性能隔离.我们基于开源SDN控制器实现了原型系统,并通过实验验证了该机制的有效性.
基金Supported by the National High-tech R&D Program("863" Program) of China (No.2013AA013505)the National Science Foundation of China(No.61472213)National Research Foundation of Korea(NRF 2014K1A1A2064649)
文摘The network resource allocation in SDN for control applications is becoming a key problem in the near future because of the conflict between the need of the flow-level flexibility control and the limited capacity of flow table.Based on the analysis of the difference of the definition of network resource between SDN and traditional IP network,the idea of the integrated allocation of link bandwidth and flow table for multiple control applications in SDN is proposed in this paper.Furthermore,a price-based joint allocation model of network resource in SDN is built by introducing the price for each of the resources,which can get the proportional fair allocation of link bandwidth and the minimum global delay at the same time.We have also designed a popular flow scheduling policy based on the proportional fair allocation of link bandwidth in order to achieve the minimum global delay.A flow scheduling module has been implemented and evaluated in Floodlight,named virtual forwarding space(VFS).VFS can not only implement the fair allocation of link bandwidth and minimum delay flow scheduling in data plane but also accelerate packet forwarding by looking up control rules in control plane.
基金supported in part by the State Major Science and Technology Special Projects under Grant No. 2012ZX03004001the National Basic Research Program (973) of China under Grants No. 2012CB315801, No. 2011CB302901the Chinese Universities Scientific Fund under Grant No. 2012RC0306
文摘In thsssse cellular network, Relay Stations (RSs) help to improve the system performance; however, little work has been done considering the fairness of RSs. In this paper, we study the cooperative game approaches for scheduling in the wireless relay networks with two-virtual-antenna array mode. After defining the metric of relay channel capacity, we form a cooperative game for scheduling and present the interpretation of three different utilization objectives physically and mathematically. Then, a Nash Bargaining Solution (NBS) is utilized for resource allocation considering the traffic load fairness for relays. After proving the existence and uniqueness of NBS in Cooperative Game (CG-NBS), we are able to resolve the resource allocation problem in the cellular relay network by the relay selection and subcarrier assignment policy and the power allocation algorithm for both RSs and UEs. Simulation results reveal that the proposed CG-NBS scheme achieves better tradeoff between relay fairness and system throughput than the conventional Maximal Rate Optimization and Maximal Minimal Fairness methods.
基金Supported by the National Science and Technology Major Project (No.2011ZX03005-004-04)the National Grand Fundamental Research 973 Program of China (No.2011CB302-905)+2 种基金the National Natural Science Foundation of China (No.61170058,61272133,and 51274202)the Research Fund for the Doctoral Program of Higher Education of China (No.20103402110041)the Suzhou Fundamental Research Project (No.SYG201143)
文摘Resource allocation is an important problem in ubiquitous network. Most of the existing resource allocation methods considering only wireless networks are not suitable for the ubiquitous network environment, and they will harm the interest of individual users with instable resource requirements. This paper considers the multi-point video surveillance scenarios in a complex network environment with both wired and wireless networks. We introduce the utility estimated by the total costs of an individual network user. The problem is studied through mathematical modeling and we propose an improved problem-specific branch-and-cut algorithm to solve it. The algorithm follows the divide-and-conquer principle and fully considers the duality feature of network selection. The experiment is conducted by simulation through C and Lingo. And it shows that compared with a centralized random allocation scheme and a cost greed allocation scheme, the proposed scheme has better per- formance of reducing the total costs by 13.0% and 30.6% respectively for the user.
基金This work was supported in part by a grant from the National Basic Research Program of China(973 Program) under Grant No.2012CB315902,the National Natural Science Foundation of China under Grant No.61379120,61170215,the Program for Zhejiang Leading Team of Science and Technology Innovation under Grant No.2011R50010-12,2011R50010-18.Zhejiang Provincial Key Laboratory of New Network Standards and Technologies (NNST)
文摘Resource scheduling algorithm for ForCES(Forwarding and Control Element Separation) networks need to meet the flexibility,programmability and scalability of node resources.DBC(Deadline Budget Constrain) algorithm relies on users select cost or time priority,then scheduling to meet the requirements of users.However,this priority strategy of users is relatively simple,and cannot adapt to dynamic change of resources,it is inevitable to reduce the QoS.In order to improve QoS,we refer to the economic model and resource scheduling model of cloud computing,use SAL(Service Level Agreement) as pricing strategy,on the basis of DBC algorithm,propose an DABP(Deadline And Budget Priority based on DBC) algorithm for ForCES networks,DABP combines both budget and time priority to scheduling.In simulation and test,we compare the task finish time and cost of DABP algorithm with DP(Deadline Priority) algorithm and BP(Budget Priority) algorithm,the analysis results show that DABP algorithm make the task complete with less cost within deadline,benifical to load balancing of ForCES networks.