软件定义网络控制器安全策略部署

针对SDN(软件定义网络)环境下的安全问题,文章提出一种在SDN控制器上部署安全策略的方案,主要实现SDN防火墙的网站过滤设计。首先设计用户访问的网站的检测过滤模块,其次通过软件设计API(应用程序编程接口)方式实现网络的安全访问控制,最后设计数据库存储拦截过滤规则和拦截日志。通过系统测试表明该安全架构具有良好的可靠性应用。     

SDN中控制器的故障切换需求分析

由多域SDNs提出解决了单控制器的可扩展问题,在可靠及可用性方面得到了一定的提升。然而在多域SDNs中,控制器的故障问题仍然存在,网络的可用、可靠性仍然无法保证。因此文章在对相关故障切换机制进行充分研究后,对故障切换进行了需求分析。     

一种考虑软件定义网络控制节点故障的控制器部署和交换机迁移方法

在软件定义网络(SDN)中,若控制节点发生不可恢复的故障,则相关交换机需向其他控制节点迁移,这将导致网络性能下降。针对这一问题,该文提出一种考虑SDN控制节点故障的控制器部署和交换机迁移方法。与现有算法仅优化交换机迁移方法不同,该方法同时考虑控制器部署位置的影响。首先利用标号传播算法构造备选子集并划分双层子集,然后在每个子集中选取合适位置部署控制器,最后为子集内节点分配相应的master和slave控制器。实验结果表明,与现有算法相比,所提方法可有效解决交换机迁移后控制器超载的问题;通过调整参数,权衡控制器故障前后网络性能,可明显改善交换机迁移后的控制链路平均时延。     

层次型多中心的SDN控制器部署

软件定义网络(SDN)通过转发与控制分离,借助控制面的集中化实现网络的灵活性和开放性.控制器部署是SDN部署运行的基础和前提.针对层次型多中心SDN的控制器部署问题,该文采用多层k路划分方法实现大规模SDN网络的区域划分,将传统的SDN多控制器直接部署转化为区域划分和域内控制器部署,同时通过减少图划分的域间割边数以降低SDN跨域流数量以提高流表构建效率.通过实验验证,较其他传统方法,该文提出的层次型多中心控制器部署方法可有效减少网络通信代价,降低流表构建代价.     

SDN中基于KMOBPSO的高可靠性控制器部署算法

针对SDN中控制器系统的单节点故障问题,兼顾系统成本和系统时延,应用N+1冗余备份模型来提高SDN控制器部署的可靠性,并将其抽象为多目标优化问题.同时,提出了一种融合K-means聚类算法和遗传算子的多目标二进制粒子群算法——KMOBPSO算法,以求解SDN控制器高可靠性部署问题的解.仿真结果表明,所提算法具有求解精度高、分布均匀、沿Pareto前沿面覆盖广的特点,能够显著提高SDN中控制器部署的可靠性.     

开源SDN控制器发展现状研究

软件定义网络（SDN）的数据平面与控制平面相分离的架构极大地简化了网络管理,将软件定义网络运用到数据中心可满足云计算动态、弹性有效使用网络的需求,同时提供自动、安全、灵活有效的资源调度能力。在SDN架构中,控制器对网络性能有着至关重要的影响,当前众多公司和科研机构均已提出了自身的控制器解决方案。首先介绍了控制器架构的设计要素,对控制器架构进行分析,之后选取业界主流的开源控制器进行详细介绍与对比分析。     

A Safe and Reliable Heterogeneous Controller Deployment Approach in SDN

Software-Defined Networking (SDN) pro-vides flexible and global network management by de-coupling control plane from data plane, and multi-ple controllers are d...     

SDN中基于分布式决策的控制器负载均衡机制

针对SDN多控制器负载均衡过程中,控制器选取僵化和交换机迁移冲突问题,提出了一种基于分布式决策的控制器负载均衡机制,分为三个阶段进行实施：首先通过周期性收集网络信息,结合控制器负载状况构建分布式迁移决策域;然后在域中依据选取概率确定迁移交换机,综合权衡数据收集、交换机迁移和状态同步三种代价选择目标控制器;最后建立迁移时钟模型,完成交换机迁移和控制器角色转换.仿真结果表明,与现有的负载均衡机制相比,降低了网络的通信开销,流建立时间平均缩短0.14s,控制器资源利用率提高了21.7%.     

SDN控制层部署策略及需求分析研究

软件定义网络(Software Defined Networking,SDN)是一种新型的网络架构,其特征包括高度解耦的控制平面与转发平面,可编程化的集中控制功能。文章介绍了SDN基本原理及控制层的关键技术,论述了SDN控制层设计的需求分析,最后针对SDN控制层部署策略做出分析与总结。     

Deadband Scheduling in Sensor Node and Controller Node for Wireless Networked Control Systems

A deadband scheduling method is proposed for wireless networked control system (WNCS), in which multiple loops share the same wireless network. The bandwidth of wireless network is very limited. The signal transmission among nodes vies for network resources, which will lead to the increase of network congestion and delays in WNCS. Deadbands are set up in both sensor and controller nodes so as to achieve network scheduling. In the sensor node, whether the sensor node sends data packet to wireless network or not depends on a deadband scheduling strategy. The deadband form is according to the difference between current sampling signal and the previous transmission signal, as well as the difference of two consecutive sampling signals. In the controller node, we use the deviations and their variation rate to build another deadband. Simulation results indicate that the proposed deadband scheduling method significantly achieves dynamic performance in WNCS, while effectively reducing network delays and network data packet traffic.     

Fault Administration by Load Balancing in Distributed SDN Controller: A Review

Wireless Personal Communications - Tremendous amount of data generated due to increasing no of users every day in the technology world. It is difficult to manage huge amount of discrete data with...     

Autonomous Domain Correlation-Based Cross-domain Network View Caching Method for SDN Distributed Controller

Software-defined networks (SDN) maintain a global view of the network,thus improving the intelligence of forwarding decisions.With the expansion of the network scale,distributed controllers are used in a variety of large-scale networks in which subnetworks managed by controller instance are called autonomous domains.We analyze statistic frequency of communication across the autonomous domain.We calculate the autonomous domain correlations for controller instances using acquired statistical information.We cache network views to highly correlated controller instances.Distributed controllers are capable of considering both the average response time and overall storage.An experiment shows that our method can fully take advantage of these two performance indicators.     

Threaded EWMA Controller Tuning and Performance Evaluation in a High-Mixed System

The exponentially weighted moving average (EWMA) controller is a very popular run-to-run (RtR) control scheme in the semiconductor industry. However, in any typical step of semiconductor process, many different products are produced on parallel tools. RtR control is usually implemented with a “threaded” control framework, i.e., different controllers are used for different combinations of tools and products. In this paper, the problem of EWMA controller tuning and performance evaluation in a mixed product system are investigated by simulation and time-series analysis. It was found that as the product frequency changed, the tuning guidelines of a threaded EWMA controller were different for different types of tool disturbances. For a stationary ARMA(1,1) noise, the tuning parameter $lambda$ should be decreased as product frequency decreases. If the tool exhibits nonstationary tool dynamics, e.g., ARIMA(1,1,1) noise, the tuning parameter should increase as the product frequency decreases.      

基于Mininet的SDN仿真与性能分析

SDN(Software Defined Network)是一种创新的网络构架,其核心理念是将控制平面和数据平面分离开来,将网络的管控集中到控制层面。OpenFlow作为南向接口标准提供了控制平面和数据平面分离的可能。文章从SDN的网络构架、SDN的硬件结构和OpenFlow协议对SDN进行了基本的分析,并采用mininet平台构建基于二层学习型交换机的简单防火墙,并对其进行仿真,交换机带宽测试值最高可达到1.09Gbps。     

WSANFlow: An Interface Protocol Between SDN Controller and End Devices for SDN-Oriented WSAN

When dealing with a Wireless Sensor and Actuator Network (WSAN) structure, one of the challenging problems is lack of flexibility in such network operations as establishment, management, and configuration. Software-defined Networking (SDN) is a promising technology for a simpler, more flexible, and less overworked network structure. Integration of SDN as a solution into the existing WSAN structures seems to be a strong candidate of deployment solutions for next generation WSAN systems. In order to get enhanced performance results for WSAN systems, we proposed an interface protocol, referred to as WSANFlow, which is responsible for all the communications between SDN controller (SDNC) and SDN-oriented end devices. The SDNC in this approach has the network intelligence and is capable of handling all the control and management operations related to the network. Thus, advanced communication operations can be managed and efficiently optimized efficiently by the SDN controller and then, subsequently, corresponding instructions can be delivered to end devices using the proposed WSANFlow protocol. In the study, we analyzed the proposed framework performance, in terms of power consumption ratio, throughput, and end to end delay metrics. Then, we compared the results with those of a ZigBee-based counterpart for different workloads such as; light, heavy and heavier load which modelizes a video stream of mild parameters. The results show that not only has the overall performance of the existing WSAN system been enhanced, but also control and management operations have been simplified by the proposed model.     

基于Mininet的SDN网络拓扑带宽性能分析

软件定义网络(SDN)是计算机网络领域的巨大创新,该技术实现了由软件来控制转发路由数据包。SDN控制器可以实现使用控制平面来管理各种虚拟交换转发设备,从而节约了大量经费,并缩短SDN的开发测试周期。由于SDN交换机种类少,并且价格昂贵,使用Mininet对计算机网络进行仿真,为SDN研究提供有效的技术支撑,其实验结果几乎可以不做任何修改直接部署到真实的硬件环境中去。     

Performance Evaluation of Optical Multicast Protection Approaches for Combined Node and Link Failure Recovery

Up to now, most research on optical multicast protection has been focused on link failure recovery, due to its predominance. Nevertheless, the impact of node failures is more severe, although they are less frequent. In a network, it might be less cost effective to implement dedicated protection approaches for separate link failure recovery and node failure recovery. A combined node and link failure recovery approach is more cost-effective and very crucial to network survivability. Hence, in this paper, we provide a detailed performance evaluation of several types of optical multicast protection approaches for 100% combined node and link failure recovery. The optical multicast protection approaches considered in this paper include the node-and-link protecting ${p}$-cycle based approach, the tree-protecting ${p}$ -cycle based approach and the optimal path pair based approach. Among them, the node-and-link protecting ${p}$-cycle based approach is newly proposed in this paper, which is a novel multicast application of previously reported ordinary ${p}$-cycle approach. Simulation results show that the integer linear programming based spare capacity optimization of node-and-link protecting ${p}$-cycle algorithm achieves the best capacity efficiency, for static traffic. For dynamic traffic, the efficiency-score based heuristic algorithm of node-and-link protecting ${p}$-cycle outperforms the other heuristic algorithms.      

应用变权模糊综合评判的目标识别效果评估

提出一种目标识别效果评估的整体思路。将变权模糊综合评判引入到目标识别效果评估，建立了一种开放的评估模型，该模型可以给出客观的、定量化的评估结果。最后，结合实测数据和拟合数据完成了仿真。     

利用简单电路提高传统PFC控制器性能

典型的两级离线PFC PFC离线功率转换器系统通常设计为两级级联型.第一级为一个升压转换器,这是因为该拓扑结构拥有连续输入电流(通过使用乘法器可实现电流波形控制)以及可实现近似单位功率因数的平均电流模式控制.但是,升压转换器需要一个比输入电压更高的输出电压,和另外一个将输出电压降压至可用电压等级的转换器(见图1).