HeS‐CoP: Heuristic switch‐controller placement scheme for distributed SDN controllers in data center networks

Software‐defined networking (SDN) has been widely researched and used to manage large‐scale networks such as data center networks (DCNs). An early stage of SDN controller experienced low responsiveness, low scalability, and low reliability. To solve these problems, distributed SDN controllers have been proposed. The concept of distributed SDN controllers distributes control messages among multiple SDN controllers. However, distributed SDN controllers must assign a master controller for each networking devices. Most previous studies, however, did not consider the characteristics of DCNs. Thus, they are not suitable to operate in DCNs. In this paper, we propose HeS‐CoP, a heuristic switch‐controller placement scheme for distributed SDN controllers in DCNs. With the control traffic load and CPU load, HeS‐CoP decides when our scheme should be performed in DCNs. To show the feasibility of HeS‐CoP, we designed and implemented an orchestrator that contains our proposed scheme and then evaluated our proposed scheme. As a result, our proposed scheme well distributes the control traffic load, decreases the average CPU load, and reduces the packet delay.     

Stateful firewall‐enabled software‐defined network with distributed controllers: A network performance study

SummarySoftware‐defined network (SDN) is constructed by decoupling the control and data plane from the forwarding devices. The control plane operations are managed by centralized or distributed controllers, and the data plane operation is managed by respective forwarding devices. SDN provides an easy and efficient management solutions for software‐programmed consolidated middlebox in virtual machines. Additionally, SDN with centralized controller faces complications like scalability, network bottle neck, and single point failure. In this study, a stateful inspection firewall acts as a middlebox in distributed SDN‐controlled network. The controller is programmed with a failure detection and recovery mechanism to provide reliability and redundancy and enhance the overall performance of the network. The objective of stateful firewall on SDN architecture is to secure the network by monitoring the current connections and maintain its state information until the connection is active. In this paper, the performance of firewall‐enabled SDN with centralized and distributed controllers are measured, compared, and analyzed. The experiments are done using POX controller, and the results are verified by Mininet network emulation tool. The results show that the stateful firewall‐enabled SDN with distributed controller network improves the security, reliability, availability, and overall performance of the network. In the proposed SDN, average network throughput is improved by 43%, average network delay is reduced by 4%, average channel utilization is increased by 40%, average network overhead is reduced by 26%, and average network response time is reduced by 23%.     

软件定义网络中可靠性优化控制器的位置研究(英)

By decoupling control plane and data plane,Software-Defined Networking(SDN) approach simplifies network management and speeds up network innovations.These benefits have led not only to prototypes,but also real SDN deployments.For wide-area SDN deployments,multiple controllers are often required,and the placement of these controllers becomes a particularly important task in the SDN context.This paper studies the problem of placing controllers in SDNs,so as to maximize the reliability of SDN control networks.We present a novel metric,called expected percentage of control path loss,to characterize the reliability of SDN control networks.We formulate the reliability-aware control placement problem,prove its NP-hardness,and examine several placement algorithms that can solve this problem.Through extensive simulations using real topologies,we show how the number of controllers and their placement influence the reliability of SDN control networks.Besides,we also found that,through strategic controller placement,the reliability of SDN control networks can be significantly improved without introducing unacceptable switch-to-controller latencies.     

Dynamic control plane management for software‐defined networks

Software‐defined network (SDN) is an emerging network paradigm that allows flexible network management by providing programmability from a separated control plane. Because of the centralized management scheme that SDN adopts, intensive control plane overhead incurs as the scale of SDN increases. The control plane overhead is mainly caused by a massive amount of control messages generated during data plane monitoring and reactive flow instantiation. By far, very few works have addressed the overhead issue on reaction flow instantiation; therefore, we mainly focus on alleviating such overhead in this work. To achieve this goal, we propose a new control plane management (CPMan) method. CPMan aims to realize the following two objectives: first, reduce the number of control messages exchanged through the control channel and second, evenly distribute the control workload across multiple controllers to mitigate the potential performance bottleneck. To realize the former, we propose a lightweight feedback loop‐based control scheme, whereas for the latter, we propose a dynamic switch‐to‐controller (DSC) placement scheme. To show the feasibility of our proposal, we implemented a prototype of the two proposed schemes on top of a carrier‐grade SDN controller and validated its performance in an emulated network. We achieved approximately 57.13% overhead reduction with feedback loop‐based control scheme, while achieved approximately 98.68% balance ratio with DSC placement scheme. Copyright © 2016 John Wiley & Sons, Ltd.     

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.     

HR-DO: High reliable–delay optimized controller placement in software-defined networking

Currently, networking technologies are pivoting towards software-defined networking (SDN) due to its easy management and centralized control system. The SDN network metrics, such as latency, scalability, robustness, and reliability largely depends on controller placement in the network domain. In view of betterment and best progression of network quality, the above mentioned network metrics need to be laid more emphases by considering controller placement. To achieve better network quality, an HR-DO (high reliable–delay optimized controller placement) is proposed. Fuzzy C-means, a meta-heuristic optimization algorithm, is used to optimize the number of controllers required in the network due to its effectiveness in cluster selection. It also eases the network latency. In addition, a minimal cut set-based Boolean logic approach and natural connectivity metric are used to improve the reliability and robustness of SDN with minimum worst-case latency. To check the performance, the proposed HR-DO is implemented on a topology-zoo network data set (OS3E and Chinanet). For comparison purposes, OCPA and K-means algorithms are taken into the account of latency, reliability, and robustness as parameters. The performance of the HR-DO approach outperforms over K-means and OCPA with respect to latency, reliability, and robustness.     

Nature‐inspired meta‐heuristic algorithms for solving the load balancing problem in the software‐defined network

The growth of the networks has difficult network management. Recently, a concept called software‐defined network (SDN) has been proposed to address this issue, which makes network management more adaptable. Control and forwarding planes are separated in SDN. The control plane is a centralized logical controller that controls the network. The forwarding plane that consists of transfer devices is responsible for transmitting packets. Because the network resources are limited, optimizing the use of resources in the networks is an important issue. Load balancing improves the balanced distribution of loads across multiple resources in order to maximize the reliability and network resources efficiency. SDN controllers can create an optimal load balancing compared to traditional networks because they have a network global view. The load‐balancing problem can be solved using many different nature‐inspired meta‐heuristic techniques because it has the NP‐complete nature. Hence, for solving load balancing problem in SDN, nature‐inspired meta‐heuristic techniques are important methods. However, to the best of our knowledge, there is not a survey or systematic review on studying these matters. Accordingly, in the area of the load balancing in the SDN, this paper reviews systematically the nature‐inspired meta‐heuristic techniques. Also, this study demonstrates advantages and disadvantages regarded of the chosen nature‐inspired meta‐heuristic techniques and considers their algorithms metrics. Moreover, to apply better load balancing techniques in the future, the important challenges of these techniques have been investigated.     

SO-CPP: Sailfish optimization-based controller placement in IoT-enabled software-defined wireless sensor networks

One of the expanding network topologies that is frequently utilized to improve network development by successfully separating the control plane and data plane is software-defined networking (SDN). In order to function inside complex sensor networks, the SDWSN system frequently relies on centralized controller logic that pulls global network information. In wireless sensor networks (WSNs), using several SDN controllers is known as a promising strategy due to reliability and performance considerations. However, using numerous controllers increases the synchronization overhead between the controllers. Consequently, it is a difficult research challenge to discover the best placement of SDN controllers to enhance the performance of a WSN, subject to the maximum number of controllers calculated based on the synchronization overhead. This research introduces a novel technique to overcome the controller placement problem (CPP) by optimizing multi-constraints within the sensor networks. For selecting the optimal controllers and placing them in an optimal location, a novel sailfish optimization (SO) strategy is introduced that can enhance the search space and maintain optimal global values throughout the iteration. Then, node clustering is performed using the fuzzy-C-means (FCM) clustering technique, which can reduce energy consumption and path delay within the network. The overall latency obtained by the proposed method is about 0.51 and 0.56 ms, and a total run time of 4 ms for both single sink and multi-sink, respectively. The proposed method is implemented in the MATLAB platform, and different performance metrics are analyzed and compared with existing techniques.     

Pluggable SDN framework for managing heterogeneous SDN networks

Software‐defined networking (SDN) is a new network paradigm that is separating the data plane and the control plane of the network, making one or more centralized controllers to supervise the behaviour of the entire network. Different types of SDN controller software exist, and research dealing with the difficulties of consistently integrating these different controller types has mostly been declared future work. In this paper, the Domino framework is proposed, a pluggable SDN framework for managing heterogeneous SDN networks. In contrast to related work, the proposed framework allows research into SDN networks controlled by different types of SDN controllers attempting to standardize the northbound API of them. Domino implements a microservice plugin architecture where users can link different SDN networks to a processing algorithm. Such an algorithm allows for, eg, adapting the flows by building a pipeline using plugins that either invoke other SDN operations or generic data processing algorithms. The Domino framework is evaluated by implementing a proof‐of‐concept implementation, which is tested on the initial requirements. It achieves the modifiability and the interoperability with an average successful exchange ratio of 99.99%. The performance requirements are met for the frequently used commands with an average response time of 0.26 seconds, and the framework can handle at least 72 plugins simultaneously depending on the available amount of RAM. The proposed framework is evaluated by means of the implementation of a shortest path routing algorithm between heterogeneous SDN networks.     

VNF-Consensus: A virtual network function for maintaining a consistent distributed software-defined network control plane

Software-defined networks (SDN) usually rely on a centralized controller, which has limited availability and scalability by definition. Although a solution is to employ a distributed control plane, the main issue with this approach is how to maintain the consistency among multiple controllers. Consistency should be achieved with as low impact on network performance as possible and should be transparent for controllers, without requiring any change of the SDN protocols. In this work, we propose VNF-Consensus, a virtual network function that implements Paxos to ensure strong consistency among controllers of a distributed control plane. In our solution, controllers can perform their control plane activities without having to execute the expensive tasks required to keep consistency. Experimental results are presented showing the cost and benefits of the proposed solution, in particular in terms of low controller overhead.     

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。     

Dynamic switch migration with noncooperative game towards control plane scalability in SDN

As software‐defined networking (SDN) is a logically centralized technology, the control plane scalability in SDN is increasingly important with the network scale increasing. Load balancing and maximizing resource utilization are very critical to the control plane in SDN, while switch migration is an effective approach to achieve these two performance metrics. However, switch migration is NP‐hard problem because it belongs to the problem of combinatorial optimization. To avoid the NP‐hard problem, we propose a switch migration scheme by adopting noncooperative game to improve the control plane scalability in SDN. First, we design a novel load balancing monitoring scheme to detect the load imbalance between controllers and trigger migrating switches. Then, we use noncooperative game among controllers to decide switch migration to get the maximizing overall profits. Last, we prove that our proposed approach can get Pareto optimality. Extensive simulations prove that our method is able to achieve a more scalable control plane with load balancing and maximizing resource utilization.     

基于SDN的多控制器部署策略的研究

随着SDN在大型网络以及广域网中的需求,如何合理、高效地部署SDN控制器,从而以较低的部署成本,获得较好的网络性能,是当前研究的热点.针对控制器数量、部署位置问题以及交换机与控制器之间的映射关系的问题,首先给出一种分布式的控制器部署方式并设定了网络相关参数,以流建立请求代价为优化目标,提出了一种基于贪心算法的控制器部署策略方案.最后通过仿真实验将所提出方案与ACL方案进行对比,证明了所提方案在性能方面有一定的提升.     

Toward manageable middleboxes in software‐defined networking

Software‐defined networking (SDN) acts as a centralized management unit, especially in a network with devices that operate under the transport layer of the OSI model. However, when a network with layer 7 middleboxes (MBs) is considered, current SDNs exhibit limitations. As such, to achieve a real‐centralized management unit, a new architecture is required that decouples the data and control planes of all network devices. In this report, we propose such a complementary architecture to the current SDN in which SDN‐enabled MBs are included along with contemporary SDN‐enabled switches. The management unit of this architecture improves network performance and reduces routing cost by considering the status of the MBs during flow forwarding. This unit consists of the following two parts: an SDN controller (SDNC) and a middlebox controller (MBC). The latter selects the best MBs for each flow and the former determines the best path according to its routing algorithm and provides information via the MBC. The results show that the proposed architecture improved performance because the utilization of all network devices including MBs is manageable.     

Performance Analysis of SDN/OpenFlow Controllers: POX Versus Floodlight

Software-Defined Networking (SDN) is an emerging network architecture that is adaptable, dynamic, cost-effective, and manageable. The SDN architecture is a form of network virtualization where the network controlling functions and forwarding functions are decoupled. A setup and configuration task of a control plane to work as an SDN controller is explained in this paper. This paper includes a brief survey of different SDN based OpenFlow-enabled controllers available in various programmable languages. This paper mainly focuses on two OpenFlow-enabled controllers, namely, POX—a Python-based controller and Floodlight—a Java-based controller. A performance comparison of both controllers is tested over different network topologies by analyzing network throughput and round-trip delay using an efficient network simulator called Mininet. A single, linear, tree and custom (user-defined) topologies are designed in Mininet by enabling external controllers. It is obtained that, a percentage improvement in round-trip time for Floodlight over POX is 11.5, 13.9, 19.6 and 14.4% for single, linear, tree and custom topology respectively. Similarly, a percentage improvement in throughput for Floodlight over POX is 5.4, 8.9, 3.8 and 4.9% for single, linear, tree and custom topology respectively.     

Energy efficient virtual network embedding for federated software‐defined networks

In this paper, we focus on energy efficient virtual network embedding in federated (multidomain) software‐defined networks (SDNs). We first formulate an optimization problem as an integer linear program (ILP) that minimizes the energy consumption of the network links, while at the same time adhering to the bandwidth and CPU requirements of the virtual network requests. We then propose a polynomial‐time heuristic algorithm, which consists of three stages. In the first stage, the top SDN controller decides on whether to partition the virtual network request into smaller subrequests and give subrequests to multiple domains or give the enitre virtual network request to a single domain, while in the second stage, each SDN controller implements virtual network embedding in its own domain. Finally, in the third stage, the algorithm performs interdomain routing if partitioning decision had been made in the first stage. Our simulation results demonstrate that our proposed algorithm yields close performance to the solutions obtained by using the optimization software CPLEX that implements our ILP.     

Survey of controller placement problem in software defined network

In order to change situation of high management complexity in current Internet,software defined network (SDN) was proposed,which mainly aimed to directly control forwarding behaviors of data-flow by using flow strategies generated by controllers.With the deployment and applications of SDN,research communities found that the controller placement in SDN network could directly affect network performance.In recent years,controller placement problem (CPP) has become a hot topic,where performance metric and searching algorithms are important research areas.Based on current researches,the existing controller placement problem was systematically analyzed and summarized,which was expected to be helpful for the follow-up research.     

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.     

DDoS attack detection and defense based on hybrid deep learning model in SDN

Software defined network (SDN) is a new kind of network technology,and the security problems are the hot topics in SDN field,such as SDN control channel security,forged service deployment and external distributed denial of service (DDoS) attacks.Aiming at DDoS attack problem of security in SDN,a DDoS attack detection method called DCNN-DSAE based on deep learning hybrid model in SDN was proposed.In this method,when a deep learning model was constructed,the input feature included 21 different types of fields extracted from the data plane and 5 extra self-designed features of distinguishing flow types.The experimental results show that the method has high accuracy,it’s better than the traditional support vector machine (SVM) and deep neural network (DNN) and other machine learning methods.At the same time,the proposed method can also shorten the processing time of classification detection.The detection model is deployed in SDN controller,and the new security policy is sent to the OpenFlow switch to achieve the defense against specific DDoS attack.     

A genetic algorithm‐based flow update scheduler for software‐defined networks

Software‐defined networking (SDN) facilitates network programmability through a central controller. It dynamically modifies the network configuration to adapt to the changes in the network. In SDN, the controller updates the network configuration through flow updates, ie, installing the flow rules in network devices. However, during the network update, improper scheduling of flow updates can lead to a number of problems including overflowing of the switch flow table memory and the link bandwidth. Another challenge is minimizing the network update completion time during large‐network updates triggered by events such as traffic engineering path updates. The existing centralized approaches do not search the solution space for flow update schedules with optimal completion time. We proposed a hybrid genetic algorithm‐based flow update scheduling method (the GA‐Flow Scheduler). By searching the solution space, the GA‐Flow Scheduler attempts to minimize the completion time of the network update without overflowing the flow table memory of the switches and the link bandwidth. It can be used in combination with other existing flow scheduling methods to improve the network performance and reduce the flow update completion time. In this paper, the GA‐Flow Scheduler is combined with a stand‐alone method called the three‐step method. Through large‐scale experiments, we show that the proposed hybrid approach could reduce the network update time and packet loss. It is concluded that the proposed GA‐Flow Scheduler provides improved performance over the stand‐alone three‐step method. Also, it handles the above‐mentioned network update problems in SDN.