Resilience and Recovery Mechanisms for Software-Defined Networking (SDN) and Cloud Networks

<p>Software defined networking (SDN) is a network framework that aims to redefine network characteristics through the programmability of network components, faster and larger network monitoring, centralized network operation, and effective detection of fraudulent traffic and special malfunctions. However, SDN networks are vulnerable to security threats that can cause complete network failure. To address this issue, in this paper, machine learning techniques are suggested for the swift detection of attacks. Various methods for detecting distributed denial of service (DDoS) attacks are evaluated, and the study identifies the most precise method for categorizing such attacks within a SDN network. The results indicate that the proposed system achieves high accuracy in detecting DDoS attacks, with ensemble learning achieving 99% accuracy. This indicates a remarkable improvement percentage in comparison to the approaches of decision tree (DT), k-nearest neighbors (KNN), and support vector machine (SVM).</p>

Software Defined Networking (SDN) paradigm brings flexibility and programmability to the network management. Deploying a set of SDN controllers to manage a Wide Area Network is challenging as it implies to determine the right number of controllers to install and their locations within the network in addition to minimize the effect of their failures. In this paper, we give a methodology that should be taken into consideration to provide a fault tolerance in SDN networks. In this paper a new model is proposed. The model is called load balancing to support fault tolerance using feedback control for SDNs (LBFTFB). LBFTFB uses feedback control theoretic techniques and uses least squares regression to find the goodness of the model. LBFTFB provides load balancing constraints for several level of backup controllers. Cascading controller failures is one of the popular controller failures. The LBFTFB model reduces the cascading failure problem effect. In order to test the efficiency of the proposed model, a reference model was used. The reference model used is the HyperFlow model. The HyperFlow model is a distributed control plane for OpenFlow. HyperFlow is an application developed on the top of the NOX controller, to enable logically centralized multi-controller architectures. Experimented were conducted to compare the performance of LBFTFB when compared to the HyperFlow. This is an ongoing research. Tentative emulation results show promising controlled results indicating that LBFTFB model outperforms the HyperFlow by 16% on average in terms of packet loss and packet delays.

Network Virtualization (NV) and Software Defined Networking (SDN) bring new business opportunities for network providers allowing them to offer the network as a service over their network infrastructure. One challenge for network providers is to provision virtual SDN services for multiple tenants while satisfying QoS and traffic engineering (TE) objectives of the higher priced SDN services, allocating the network resources cost effectively and managing the virtual services that span over customer premises, wide area and cloud domains. In this paper, we describe an algorithmic framework that includes QoS management and traffic-engineering methods for managing virtual services over large SDN networks. The algorithmic framework is part of an intelligent service management architecture which employs optimization algorithms, machine learning and artificial intelligence techniques into the management of virtual services over SDN networks.

Software Defined Networking (SDN) revolutionizes network control by separating the control plane from the data plane. Although the latter improves SDN agility and scalability, it creates a security hole, particularly in a central control plane, leading to SDN environments becoming high-profile targets for advanced cybersecurity threats. Due to static and signature-based point-in-time behavior, traditional security methods are unable to keep up with modern attacks that are an anomaly to SDNs. Artificial Intelligence (AI) with its different applications and techniques, has the capability of detecting SDN cyber threats’ anomalies. This paper presents the results of a literature scoping exercise that used a total of 54 papers that looked at AI-driven anomaly detection in SDN. The findings showed that control theory, activity theory, and anomaly detection theory are three theoretical aspects that contribute to the topic of AI-driven anomaly detection in SDN. Furthermore, different machine learning algorithms give different results. In this regard, Random Forest (RF), Support Vector Machine (SVM), and Multi-Layer Perceptron would help in detecting threats of a familiar nature, while autoencoders and K-means can detect unfamiliar threats. While deep learning architectures such as Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNN) support low-latency anomaly detection while maintaining throughput and network stability. The findings could be the basis of providing a conceptual framework on how an intelligent, adaptive, and resilient SDN with real-time threat defense mechanisms could be designed, developed, and deployed.

Cloud and parallel computing are fundamental components in the processing of large data sets. Deployments of distributed computers require network infrastructure that is fast, efficient, and secure. Software Defined Networking (SDN) separates the forwarding of network data by switches (data plane) from the setting and managing of network policies (control plane). While this separation provides flexibility for setting network policies affecting the establishment of network flows in the data plane, it provides little to no fault tolerance for failures, either benign or caused by corrupted/malicious applications. Such failures can cause network flows to be incorrectly routed through the network or stop such flows altogether. Without protection against faults, cloud network providers using SDN run the risk of inefficient allocation of network resources or even data loss. Furthermore, the asynchronous nature existing protocols for SDN does not provide a mechanism for consistency in network policy updates across multiple switches.In addition, cloud and parallel applications require an efficient means for accessing local system data (input data sets, temporary storage locations, etc.). While in many cases it may be possible for a process to access this data by making calls directly to a file system (FS) kernel driver, this is not always possible (e.g. when using experimental distributed FSs where the needed libraries for accessing the FS only exist in user space).This dissertation provides a design for fault tolerance of SDN and infrastructure for advancing the performance of user space FSs. It is divided into three main parts. The first part describes a fault tolerant, distributed SDN control plane framework. The second part expands upon the fault tolerant approach to SDN control plane by providing a practical means for dynamic control plane membership as well as providing a simple mechanism for controller authentication through threshold signatures. The third part describes an efficient framework for user space FS access.This research makes three contributions. First, the design, specification, implementation, and evaluation of a method for fault tolerant SDN control plane that is inter-operable with existing control plane applications involving minimal instrumentation of the data plane runtime. Second, the design, specification, implementation and evaluation of a mechanism for dynamic SDN control plane membership that all ensure consistency of network policy updates and minimizes switch overhead through the use of distributed key generation and threshold signatures. Third, the design, specification, implementation, and evaluation of a user space FS access framework that is correct to the Portable Operating System Interface (POSIX) specification with significantly better performance over existing user space access methods, while requiring no implementation changes for application programmers.

In recent days, cloud computing has become an outstanding technique in all areas that provide various computing resources. The specialty of cloud computing is that it provides excellent services in monitoring, communication, software platform, and infrastructure. Basically, cloud network limits its services in a specific location on the particular zone. When the targeted area is near, the specific server completes the request generated from the user. Few servers are busy while some other servers are idle when it comes to the services. In this case, the provided resources are not used efficiently. To overcome this problem, this paper proposes a concept known as flow-based dynamic load balancing algorithm in cloud networking with software defined networks (SDN). The fundamental principle of this proposed algorithm is distributing client requests equally to all the available servers in a specific cloud network. The main aim of the load balancing algorithm is to optimise the resource utilisation by increasing the throughput, reduce flow completion time, decrease the response time and eliminate the overloading of any individual resource. SDN is mainly used in applying various software functionalities into hardware devices such as routers, switches, etc.

Hybrid software-defined networking (SDN) networks are intermediates between traditional networks and SDN networks. However, since the hybrid control planes are structurally different from the single control plane, updating the hybrid SDN networks lacks effective techniques. In this paper, by utilizing the segment routing (SR) mechanism to update hybrid SDN networks and formalize the application of SR during updates, we propose a cooperative consistency update algorithm that can coordinate SR, two-phase commit techniques. The algorithm is capable of preserving consistency with low overhead and excellent applicability. When a network requires updating, the algorithm leverages the SR mechanism to stitch a new path using the available path segments; thus, flows can use the new path immediately. If the update cannot be completed with the SR mechanism alone, the algorithm can utilize the two-phase commit technique to update the nodes through duplication. Notably, our experiments based on real topologies indicate that, compared to prior techniques, the proposed algorithm can greatly reduce the TCAM overhead.

It is Software defined networking (SDN) is a new network approach which allows separation of the forwarding plane from the control plane in network architecture, that aids in the efficient control of data flow and gives network administrators a software- based method for managing and configuring the networks. In contrast to SDN, which provides a centralized controller, traditional network topologies require the individual configuration of each network device. Due to a centralized control functionality, SDN is increasingly used by businesses, telecom service providers, cloud providers, and many other users, as it encourages flexibility and scalability in the network while also assisting in streamlining operations. There is significant demand of SDN for effective and adaptable networks. SDN adoption has also been rising in the campuses and corporate sectors. This paper provides a concise overview of the Software Defined Networks (SDN) topic, which is essential for comprehending the fundamental principles of SDN in order to conduct research and implement SDN in many sectors. The initial focus is on the discussion of traditional networks, including their limitations, as well as the many issues encountered by industries. The topic of Software Defined Networks is explored, encompassing their necessity, historical background, architectural structure, interfaces, and models. Next, the many applications of Software-Defined Networking (SDN) are examined. Additionally, the various SDN case studies, vendors, setup issues with SDN, and current research fields were elucidated. The discussion focuses on the role of Software-Defined Networking (SDN) in Artificial Intelligence (AI), specifically addressing the issues associated with implementing SDN in AI.

Software Defined Networking (SDN) is a new network design paradigm. SDNs depend on decoupling thecontrol plane from the forwarding plane atthe network switches. SDNs use logically centralized controllers that are responsible for directing the forwarding decisions to the connected switches. In this paper, we design a new model to provide fault tolerance using feedback control to SDNs based on load balancing schemes. The model is called load balancing to support fault tolerance using feedback control for SDNs (LBFTFB). LBFTFB improves the resistance to cascaded failures. Cascading failures cause an enormous increase in the load of the surrounding controllers as a subsequence of the failure of a certain controller. On SDN controller failures, the LBFTFB distributes the load on the failed controller among thesurrounding controllers based on a defined feedback control policy. The LBFTFB model reduces the cascading failure problem when compared to HyperFlow model which is considered as a reference model. The HyperFlow model is a distributed control plane for OpenFlow. HyperFlow model suffers from a dramatic increase in the load of the surrounding controllers as a consequence of the failure of a certain controller. This is an ongoing research. Tentative emulation results are promising and show that the LBFTFB model outperforms the reference model for cascading failures by 14% in terms of packet loss and 17% in terms of packet delays

The term software defined networking (SDN) is a network model that contributes to redefining the network characteristics by making the components of this network programmable, monitoring the network faster and larger, operating with the networks from a central location, as well as the possibility of detecting fraudulent traffic and detecting special malfunctions in a simple and effective way. In addition, it is the land of many security threats that lead to the complete suspension of this network. To mitigate this attack this paper based on the use of machine learning techniques contribute to the rapid detection of these attacks and methods were evaluated detecting DDoS attacks and choosing the optimum accuracy for classifying these types within the SDN, the results showed that the proposed system provides the better results of accuracy to detect the DDos attack in SDN network as 99.90% accuracy of Decision Tree (DT) algorithm.

Computer network conditions that are dynamic and also complex make network operators often make mistakes in the configuration of a network device. However, these mistakes can be resolved by the emergence of a centralized programming approach which is now known as Software Defined Network (SDN). In addition, with the development of information technology, several services in computer networks also grabbed a lot of attention from the world, such as video streaming services. A video streaming service that is so massive is capable of having an impact on data traffic on computer networks. In SDN networks, controllers are a core component of this architecture. Currently, many different types of SDN controllers can be used, however, choosing a controller in an SDN network must be an important concern. This is related to the performance of the SDN controller which is a requirement for SDN network development. This research will analyze the performance of the video streaming service using the Real-time Transport Protocol (RTP) streaming protocol in testing two SDN controllers, namely Pox and Opendaylight. Based on the tests that have been carried out, it produces Quality of Service (QoS) values in the form of throughput, delay, and jitter. The Pox controller excels in the value of QoS delay and throughput. As for the jitter value, both Pox and Opendaylight controllers have good indexes.

Popularity of Software Defined Network (SDN) technology is spreading rapidly and corporate networks are deploying to take advantages of its features. Control & Data planes are segregated in SDN and simple forwarding devices enabled with OpenFlow protocol is used to forward data planes. The responsibility of control planes are taken by controller into SDN network. The data packets are generated by source host and forwarded to destination host by seeing destination address of such packet without validating source host address. Flow of data packets is controlled by controller setting up required flow entries into flow table of forwarding devices. Initially forwarding devices do not have any security & control and they are not able to forward traffic generated by attached hosts. Data packets of attached hosts are sent to the controller by forwarding devices for generation of control packets and setup necessary flow entries into device flow table. Source IP spoofed attacks may be setup by attacker in this situation and perform different type of attacks. There are some techniques for prevention and mitigation from such attacks. In this study, we offer an analysis of available source address validation techniques for SDN environment. The study discussed various features, limitations and gaps of source address validation solutions and provided various areas in SDN Security for future research.

Considering the Internet of Things (IoT) impact in today’s world, uninterrupted service is essential, and recovery has received more attention than ever before. Fault-tolerance (FT) is an essential aspect of network resilience. Fault-tolerance mechanisms are required to ensure high availability and high reliability in systems. The advent of software-defined networking (SDN) in the IoT plays a significant role in providing a reliable communication platform. This paper proposes a data plane fault-tolerant architecture using the concepts of software-defined networks for IoT environments. In this work, a mathematical model called Shared Risk Link Group (SRLG) calculates redundant paths as the primary and backup non-overlapping paths between network equipment. In addition to the fault tolerance, service quality was considered in the proposed schemes. Putting the percentage of link bandwidth usage and the rate of link delay in calculating link costs makes it possible to calculate two completely non-overlapping paths with the best condition. We compare our two proposed dynamic schemes with the hybrid disjoint paths (Hybrid_DP) method and our previous work. IoT developments, wireless and wired equipment are now used in many industrial and commercial applications, so the proposed hybrid dynamic method supports both wired and wireless devices; furthermore multiple link failures will be supported in the two proposed dynamic schemes. Simulation results indicate that, while reducing the error recovery time, the two proposed dynamic designs lead to improved service quality parameters such as packet loss and delay compared to the Hybrid_DP method. The results show that in case of a link failure in the network, the proposed hybrid dynamic scheme’s recovery time is approximately 12 ms. Furthermore, in the proposed hybrid dynamic scheme, on average, the recovery time, the packet loss, and the delay improved by 22.39%, 8.2%, 5.66%, compared to the Hybrid_DP method, respectively.

This book introduces two basic big data processing paradigms for batch data and streaming data. Representative programming frameworks are also presented, as well as software defined networking (SDN) and network function virtualization (NFV) technologies as key cloud networking technologies. The authors illustrate that SDN and NFV can be applied to benefit the big data processing by proposing a cloud networking framework. Based on the framework, two case studies examine how to improve the cost efficiency of big data processing. Cloud Networking for Big Data targets professionals and researchers working in big data, networks, wireless communicationsand information technology. Advanced-level students studying computer science and electrical engineering will also find this book valuable as a study guide.

Accurate traffic and latency measurements are the foundation of many network management tasks, including traffic engineering and network provision, network diagnosis and fault recovery. However, SNMP-based traffic measurement techniques in traditional IP networks only provide measurements for link loads, which cannot lead to accurate measurements. For link latency measurement, it has been shown that a traditional IP network needs to deploy a large number of monitors to achieve accurate measurements. The software-defined network (SDN) architecture opens up new opportunities to tackle the traffic matrix estimation and latency measurement problems. For the traffic estimation problem, the counter in a flow table entry of an SDN switch can directly measure the size of a matched flow. This dramatically increases the number and type of flows that can be measured accurately. For link latency measurement, SDN switches provide flexible routing and hence allow even non-shortest paths to be measured in the network. This routing flexibility can reduce the number of monitors deployed in the network to achieve a complete measurement of all the link latencies. But the additional measurement capabilities of SDNs also have limitations: the flow table entries added for measurements must be limited due to the high cost of ternary content-addressable memory (TCAM), and the amount of additional traffic for measurement should also be confined. Under these constraints, this research investigates the traffic matrix estimation and link latency measurement problems for a pure SDN network and a hybrid SDN network. A hybrid network, which consists of SDN switches and conventional routers, will exist for some time to come since it is economically impossible to replace every conventional router overnight. For the traffic matrix estimation problem, we propose a new estimation framework, which can be used in both SDNs and hybrid SDNs. Performance evaluation results show that the new framework can achieve highly accurate traffic measurement with a small number of TCAM entries. For the link latency measurement problem, we first present minimum monitor placement schemes in hybrid SDNs. It has been shown that our schemes not only reduce the number of required monitors but also guarantee full link measurements even when link weights change. We then present two time-stamp-based link latency measurement schemes for pure SDNs and show that the proposed schemes can minimize the number of consumed TCAM entries and the amount of traffic generated for measurement.