软件定义网络中基于效率区间的负载均衡在线优化算法

在大型复杂软件定义网络中,为提高网络负载均衡,减少控制器与交换机间的传播时延,该文提出一种基于效率区间的负载均衡在线优化算法。在初始静态网络中,通过贪心算法选择初始控制器集合,并以其为根节点构建M棵改进代价的最小生成树(MST),确定初始M个负载均衡的子网;当网络流量发生变化时,通过广度优先搜索(BFS)调整子网间交换机映射关系使其满足效率区间,保证任意时刻网络的负载均衡。算法均以网络连通性为基础,且均以传播时延为目标重新更新控制器集合。仿真实验表明,该算法在保证任意时刻网络负载均衡的同时,可以保证较低的传播时延,与Pareto模拟退火算法、改进的K-Means算法等相比,可以使网络负载均衡情况平均提高40.65%。     

SDN中基于Q-learning的动态交换机迁移算法

由于网络流量动态变化,控制器负载均衡成为大规模部署软件定义网络研究的重点.提出基于Q-learning的动态交换机迁移算法,首先对软件定义网络中的控制器部署问题建模,再应用Q-learning反馈机制学习实时网络流量,最后根据Q表格将交换机从高负载控制器动态迁移到低负载控制器上,实现控制器的负载均衡.仿真结果表明,所提算法能够获得较低的控制器负载标准方差.     

基于迁移演员-评论家学习的服务功能链部署算法

针对5G网络切片环境下由于业务请求的随机性和未知性导致的资源分配不合理从而引起的系统高时延问题,该文提出了一种基于迁移演员-评论家(A-C)学习的服务功能链(SFC)部署算法(TACA)。首先,该算法建立基于虚拟网络功能放置、计算资源、链路带宽资源和前传网络资源联合分配的端到端时延最小化模型,并将其转化为离散时间马尔可夫决策过程(MDP)。而后,在该MDP中采用A-C学习算法与环境进行不断交互动态调整SFC部署策略,优化端到端时延。进一步,为了实现并加速该A-C算法在其他相似目标任务中(如业务请求到达率普遍更高)的收敛过程,采用迁移A-C学习算法实现利用源任务学习的SFC部署知识快速寻找目标任务中的部署策略。仿真结果表明,该文所提算法能够减小且稳定SFC业务数据包的队列积压,优化系统端到端时延,并提高资源利用率。     

一种基于负载均衡的高能效LLN路由算法

针对低功耗有损网络(LLN)中由于节点部署不均匀易导致负载不均衡的问题,提出一种基于负载均衡的高能效LLN路由算法(EELB-RPL)。通过将链路质量、传输时延、节点剩余能量、吞吐量以及拥塞检测因子等因素相结合选出最优父节点;再通过调整溪流计时器,使节点根据网络拓扑密度调整自身抑制机制,避免了抑制不公平性导致负载不均衡。仿真结果表明,所提算法与现有算法相比较,能够有效实现负载均衡。其中,数据包投递率提升了14.6%,根节点吞吐量提升了28.5%,网络平均寿命提升了8.96%。     

基于改进遗传算法的SDN控制器放置优化研究

与传统的网络不同,软件定义网络将网络的控制层与数据层分开,从而提供了网络的集中视图,可以灵活控制网络流量。然而在大规模的软件定义网络中不合理的控制器部署会直接影响到网络的性能。为了找到网络中最优的控制器放置位置和数量,同时最小化时延、均衡负载并优化成本,提出一种基于改进遗传算法的求解算法。在该算法中引入贪婪思想并改进变异算子从而改善其收敛速度和寻优能力。此外,在一些真实的拓扑上将该算法与其他求解算法进行对比,结果表明改进的遗传算法性能更好。     

改进NSGA-Ⅲ的卫星网络多控制器部署方法

软件定义卫星网络能够促使卫星网络管理便利化、网络控制灵活化、网络配置简单化,通过卫星控制器节点可实时掌握卫星网络的状态信息,实现卫星网络资源的集中控制与高效分配。针对卫星网络控制器部署问题研究中,星间链路的可靠性常常被忽略的问题,建立了以降低控制链路时延、提升星间链路的可靠性、维持负载均衡为研究目标,提出了一种改进的非支配排序遗传算法-Ⅲ（Improved Non-dominated Sorting Genetic AlgorithmⅢ,INSGA-Ⅲ）的卫星网络多控制器部署方法。仿真结果表明,基于INSGA-Ⅲ算法的多控制器部署方法能够降低网络时延,同时在星间链路的可靠性和网络负载均衡方面均有显著提升,在时延、负载均衡、可靠性方面,INSGA-Ⅲ相比于INSGA-Ⅱ分别提升约10.12%、97.37%、0.50%,INSGA-Ⅲ相比于NSGA-Ⅲ分别提升约6.25%、95.49%、0.44%。     

软件定义网络中基于负载均衡的多控制器部署算法

随着软件定义网络规模扩大,控制层与数据层解耦带来了诸如控制器部署等新问题。该文提出基于负载均衡的多控制器部署算法(Multi-Controller Deployment Algorithm Based on Load Balance, MCDALB)。算法首先根据网络拓扑结构及其负载情况,确定控制器数量K;然后根据控制器容量限制,提出一种近似比为2的多控制器负载均衡算法,将网络划分成K个控制区域;最后根据区域内所有交换机到控制器距离总和最小原则,在控制区域部署控制器。为了验证算法的性能,选取实际网络拓扑进行实验。实验结果表明,与AL, WL算法相比,该算法在满足控制器负载近似比为2的同时,网络最大延时差距不超过0.65 ms。     

基于SDN的负载均衡动态流量调整策略

在基于SDN的数据中心中,负载均衡问题一直是研究的热点问题。文中利用SDN控制器集中控制网络状态的特点,提出了一种动态流量调整(DFA)策略以达到网络负载均衡的目的。通过获得网络的实时状态,定义负载平衡度作为触发策略的阈值,可以在流量传输过程中动态调整传输路径。文中对该策略的实施过程进行了详细描述,通过仿真平台进行了实验。在不同负载情况下,对比实验结果表明,实施该策略的网络中心在平均时延、丢包率等方面均有更好的表现,网络整体性能提高了5%～10%。     

在线SDN控制器性能监控器设计与研究

随着网络规模和网络数据流量的变大,软件定义网络控制器性能问题可能成为网络瓶颈,为了在动态负载情况下仍然可以获取控制器准确的性能状态,设计了一个在线的性能监控器框架,并定义了控制器时延性能衡量指标Lpdex.在一款广泛使用的FloodLight控制器中实现了性能监控器的简单原型,经过实验评估,验证了控制器性能监控器的可行性和Lpdex可在线准确地反映控制器在不同负载情况下时延性能的变化.     

5G网络切片场景中基于预测的虚拟网络功能动态部署算法

针对无线虚拟化网络在时间域上业务请求的动态变化和信息反馈时延导致虚拟资源分配的不合理,该文提出一种基于长短时记忆(LSTM)网络的流量感知算法,该算法通过服务功能链(SFC)的历史队列信息来预测未来负载状态。基于预测的结果,联合考虑虚拟网络功能(VNF)的调度问题和相应的计算资源分配问题,提出一种基于最大最小蚁群算法(MMACA)的虚拟网络功能动态部署方法,在满足未来队列不溢出的最低资源需求的前提下,采用按需分配的方式最大化计算资源利用率。仿真结果表明,该文提出的基于LSTM神经网络预测模型能够获得很好的预测效果,实现了网络的在线监测;基于MMACA的VNF部署方法有效降低了比特丢失率的同时也降低了整体VNF调度产生的平均端到端时延。     

Multi-controller balancing deployment strategy based on reliability evaluation in SDN

The control plane has low reliability and controller loads were unbalanced in distributed software defined networks,and a multi-controller balancing deployment strategy based on reliability evaluation was proposed.Firstly,node reliability was evaluated by weighting node efficiency and path quality,optimizing the controller location with balancing factor.Then based on improved k-center clustering,the redundant functions was introduced and the switch allocation was completed according to node attractiveness degree and controller load balancing rate,achieving a reasonable SDN sub-domain planning.The simulations show that compared with the existing strategy the number of required controllers is reduced by an average of 22.1%.The control plane elasticity is enhanced,and the controller load balancing performance has been improved significantly.     

Balanced multiple controllers placement with latency and capacity bound in software-defined network

Software-defined network (SDN) used a network architecture which separates the control plane and data plane. The control logic of SDN was implemented by the controller. Because controller's capacity was limited, in large scale SDN networks, single controller can not satisfy the requirement of all switches. Multiple controllers were needed to han-dle all data flows. By the reason that the latency between controller and switch would significantly affect the forwarding of new data flow, the rational placement of controllers would effectively improve the performance of entire network. By partition the network into multiple sub domains, on the base of spectral clustering, a method that added a balanced de-ployment object function into k-means was given and a balanced multiple controllers placement algorithm in SDN net-works which has the latency and capacity limitations was proposed. In this approach, a penalty function was introduced in the algorithm to avoid isolation nodes appearing. The simulations show that this algorithm can balance partition the net-work, keep the latency between controller and switch small and keep loads balancing between controllers.     

On the use of the genetic programming for balanced load distribution in software-defined networks

As a new networking paradigm, Software-Defined Networking (SDN)enables us to cope with the limitations of traditional networks. SDN uses a controller that has a global view of the network and switch devices which act as packet forwarding hardware, known as “OpenFlow switches”. Since load balancing service is essential to distribute workload across servers in data centers, we propose an effective load balancing scheme in SDN, using a genetic programming approach, called Genetic Programming based Load Balancing (GPLB). We formulate the problem to find a path: 1) with the best bottleneck switch which has the lowest capacity within bottleneck switches of each path, 2) with the shortest path, and 3) requiring the less possible operations. For the purpose of choosing the real-time least loaded path, GPLB immediately calculates the integrated load of paths based on the information that receives from the SDN controller. Hence, in this design, the controller sends the load information of each path to the load balancing algorithm periodically and then the load balancing algorithm returns a least loaded path to the controller. In this paper, we use the Mininet emulator and the OpenDaylight controller to evaluate the effectiveness of the GPLB. The simulative study of the GPLB shows that there is a big improvement in performance metrics and the latency and the jitter are minimized. The GPLB also has the maximum throughput in comparison with related works and has performed better in the heavy traffic situation. The results show that our model stands smartly while not increasing further overhead.     

An adaptive load-aware routing algorithm for multi-interface wireless mesh networks

In wireless mesh networks, the number of gateway nodes are limited, when the nodes access to the internet by fixed gateway node, different requirements of nodes lead to the dataflow shows heterogeneity. Many new routing metrics and algorithms existing in traditional wired networks and the Ad Hoc network, can not be directly applied to wireless mesh networks, so how to design a routing metric and algorithm which can dynamically adapt to current networks topology and dataflow changes, avoid bottleneck node, and select the most stable and least congestion link to establish a route is very important. In this paper, we presented a new dynamic adaptive channel load-aware metric (LAM) to solve the link load imbalance caused by inter-flow and inner-flow interference, designed a self-adaptive dynamic load balancing on-demand routing algorithm through extending and improving AODV routing method with the LAM, to achieve flow balance, reduce the high packet loss ratio and latency because congestion and Packet retransmission, and can increase Network Throughput.     

Active overload prevention based adaptive MAP selection in HMIPv6 networks

Multi-level mobile anchor points (MAP) architecture is deployed in large-scale wireless/mobile networks using HMIPv6 to achieve better mobility service, while selecting the most suitable serving MAP for the mobile nodes (MNs) to enhance the whole network performance has been a critical issue. An adaptive MAP selection based on active overload prevention (MAP-AOP) hence is proposed. The MAP periodically evaluates the load status by using dynamic weighted load evaluation algorithm, and then sends the load information to the covered access routers (AR) by using the expanded routing advertisement message in a dynamic manner. Taking achieving the load balancing among the available MAPs, the current serving AR executes the active overload prevention to select MAP candidates for the MN pending a handover, and then adaptively selects an optimal one from the candidates by comprehensively considering the system cost and the average handover latency caused by each candidate. The simulation conducted on the NS-2 platform indicates that MAP-AOP outperforms the comparative MAP selection schemes with the optimized system cost and average handover latency, and better load balancing.     

Power control for data load balancing with coverage in dynamic femtocell networks

We consider the load balancing and coverage problem of femtocell networks in indoor environment. We propose a novel framework exploiting the Voronoi diagram with respect to the radio propagation distance. Our initial power assignment scheme achieves maximal indoor coverage with minimized maximum required transmit power, which results in reduced interference. Our approach can adopt any radio propagation model to achieve more accurate coverage for an indoor environment with various obstacles. Time varying data traffic may cause unbalanced data load of base stations producing traffic overload. Our dynamic power control algorithm redistributes the data load by automatically adjusting transmit power levels according to data traffic estimation while preserving coverage. Moreover, we present an algorithm to cope with the dynamic deletion and insertion of femto base stations. Simulations show that the proposed scheme achieves better coverage, reduced interference, and good load balance compared to previous algorithms.     

多模车载视频监控系统中负载均衡算法设计

为了提高车载视频监控系统的视频传输质量,本文基于排队论,利用多链路捆绑技术,设计了基于TD-SCDMA,WCDMA和CDMA2000的多模车载视频监控的负载均衡算法.该负载均衡算法以各个网卡在数据传输时的传输时延、信号强度、运营商因子和丢包率加权作为均衡因子,依次作为负载均衡时数据转发的依据.经实验测试,本算法能够充分利用增加用户卡来增加传输速率,并通过捆绑不同运营商的用户卡来提高数据传输的稳定性,并且该算法较常用的轮询算法的数据传输速率有了较大提高.     

传感网中能量均衡高效的源位置隐私保护协议

针对无线传感网中现有信源匿名协议大都存在无法同时兼顾源位置匿名性、延迟和生存周期的问题,对数据混淆、虚假信息注入等方法在隐私性、网络性能等方面进行了分析,在此基础上首次提出了基于匿名量化动态混淆环的源位置隐私保护协议ADRing：以能量均衡消耗确保的动态混淆环以及相应基于区位象限的匿名度量化评估机制来高效地保护信源节点位置隐私。仿真实验结果表明,ADRing相比于现有协议能明显改善网络能耗和延迟均衡性,延长网络寿命,能满足网络节点对于数据分组匿名性和延迟的不同需求。     

可重配置系统中的联合负载控制及其终端选择算法

该文基于现有端到端可重配置系统架构,提出了一种改进的动态门限联合负载控制方法,以适应不同负载条件下对负载均衡的要求,达到资源的有效利用。同时,结合终端的可重配置特点及业务要求的多样性,在执行网络发起的垂直切换过程中采用了基于层次分析法的终端选择算法,从而在均衡异构网络负载的同时减少系统开销和对用户的影响。仿真结果表明,该方法能够在保证用户满意度的同时有效地减轻重叠覆盖区域的网络拥塞,提高系统性能。