Today, data centers host a wide variety of applications which generate a mix of diverse internal data center traffic. In a data center environment 90% of the traffic flows, though they constitute only 10% of the data carried around, are short flows with sizes up to a maximum of 1MB. The rest 10% constitute long flows with sizes in the range of 1MB to 1GB. Throughput matters for the long flows whereas short flows are latency sensitive. This thesis studies various data center transport mechanisms aimed at either improving flow completion time for short flows or throughput for long flows. Thesis puts forth two data center transport mechanisms: (1) for improving flow completion time for short flows (2) for improving throughput for long flows. The first data center transport mechanism proposed in this thesis, FA-DCTCP (Flow Aware DCTCP), is based on Data Center Transmission Control Protocol (DCTCP). DCTCP is a Transmission Control Protocol (TCP) variant for data centers pioneered by Microsoft, which is being deployed widely in data centers today. DCTCP congestion control algorithm treats short flows and long flows equally. This thesis demonstrate that, treating them differently by reducing the congestion window for short flows at a lower rate compared to long flows, at the onset of congestion, 99th percentile of flow completion time for short flows could be improved by up to 32.5%, thereby reducing their tail latency by up to 32.5%. As per data center traffic measurement studies, data center internal traffic often exhibit predefined patterns with respect to the traffic flow mix. The second data center transport mechanism proposed in this thesis shows that, insights into the internal data center traffic composition could be leveraged to achieve better throughput for long flows. The mechanism for the same is implemented by adopting the Software Defined Networking paradigm, which offers the ability to dynamically adapt network configuration parameters based on network observations. The proposed solution achieves up to 22% improvement in long flow throughput, by dynamically adjusting network element’s QoS configurations, based on the observed traffic pattern.

In mobile cloud computing (MCC), rich mobile application data is processed at the cloud infrastructure by reliving resource limited mobile devices from computationally complex tasks. However, due to the ubiquitous and mobility nature, providing time critical rich applications over remote cloud infrastructure is a challenging task for mobile application service providers. Therefore, according to the literature, close proximity placement of cloud services has been identified as a way to achieve lower end-to-end access delay and thereby provide a higher quality of experience (QoE) for rich mobile application users. However, providing a higher Quality of Service (QoS) with mobility is still a challenge within close proximity clouds. Access delay to a closely placed cloud tends to be increased over time when users move away from the cloud. However, reactive resource relocation mechanism proposed in literature does not provide a comprehensive mechanism to guarantee the QoS and as well as to minimize service provisioning cost for mobile cloud service providers. As a result, using the benefits of SDN and the data plane programmability with logically centralized controllers, a resource allocation framework was proposed for IaaS mobile clouds with regional datacenters. The user mobility problem was analyzed within SDN-enabled wireless networks and addressed the possible service level agreement violations that could occur with inter-regional mobility. The proposed framework is composed of an optimization algorithm to provide seamless cloud service during user mobility. Further a service provisioning cost minimization criteria was considered during an event of resource allocation and inter-regional user mobility.

In today’s scenario, energy efficiency has become one of the most crucial issues for Data Center Networks (DCN). This paper analyses the energy saving capability of a Data center network using Segment Routing (SR) based model within a Software Defined Network (SDN) architecture. Energy efficiency is measured in terms of number of links turned off and for how long the links remain in sleep mode. Apart from saving the energy by turning off links, our work further efficiently manages the traffic within the available links by using Per-packet based load balancing approach. Aiming to avoid congestion within DCN’s and increase the sleeping time of inactive links. An algorithm for deciding the particular set of links to be turned off within a network is presented. With the introduction of per-packet approach within SR/SDN model, we have successfully saved 21 % of energy within DCN topology. Results show that the proposed Per-packet SR model using Random Packet Spraying (RPS) saves more energy and provides better performance as compared to Per-flow based SR model, which uses Equal Cost Multiple Path (ECMP) for load balancing. But, certain problems also come into picture using per-packet approach, like out of order packets and longer end to end delay. To further solidify the effect of SR in saving energy within DCN and avoid previously introduced problems, we have used per-flow based Flow Reservation approach along with a proposed Flow Scheduling Algorithm. Flow rate of all incoming flows can be deduced using Flow reservation approach, which is further used by Flow Scheduling Algorithm to increase Bandwidth utilization Ratio of links. Ultimately, managing the traffic more efficiently and increasing the sleeping time of links, leading to more energy savings. Results show that, the energy savings are almost similar in per-packet based approach and per-flow based approach with bandwidth reservation. Except, the average sleeping time of links in per-flow based approach with bandwidth reservation decreases less severely as compared to per-packet based approach, as overall traffic load increases.

This study aims to examine the effects of transforming a monolithic server system into a microservice architecture, focusing on the increased latency introduced by using a microservice orchestrator. The microservice orchestrator was implemented using an OpenFlow switch controlled by the Beacon and Ryu OpenFlow controllers. These controllers, along with the round robin, random assign and a server-aware load balancing algorithm, were all compared in order to find the combination resulting in the lowest latency and highest achieved server balance in varying network environments. We show that the OpenFlow switch enforces a client-aware load balancing policy and that only the initial request is handled by the controller, effectively reducing the importance of choosing the optimal OpenFlow controller. In addition, the round robin load balancer was preferred when dealing with homogeneous requests, and a server-aware load balancer was required for heterogeneous requests. For most requests, the system would only slow down by a few microseconds using the proposed architecture. However, for 0.001\% of all requests, the slowdown was much more significant, with each of those requests being at least 100 times slower than when using a monolithic server architecture.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第22809号 / 情博第739号 / 新制||情||126(附属図書館) / 京都大学大学院情報学研究科通信情報システム専攻 / (主査)教授 大木 英司, 教授 守倉 正博, 教授 原田 博司 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

This thesis covers utilization of software defined networks for lawful interception purposes. Based on specific implementation of lawful interception system SLIS developed by Sec6Net group, suggests improvements aiming at more precise identification of intercepted users and better effectivity of system resources. First aim is achieved by implementation of a new module for dynamic identification component while the other one alters configuration mechanism for probes and OpenFlow switches.

No description available.

The notion of the networked society is more than ever true nowadays. The Internet has a big impact on our daily lives. Network operators provide the underlying infrastructure and continuously deploy services in order to meet customer demands. The amount of data transported through operator networks is also increasing with the introduction of new high band width services and over the network content. That being said, operators, most often deploy or operate networks to meet these demands without any regard to energy-efficiency. As the price of electricity continues to grow,  tends to become a problem with serious implications. To solve this problem a trend towards more energy efficient networks has emerged. In this thesis, we investigate a way to facilitate the introduction of new energy efficiency paradigms for fixed networks. Towards this end, we investigate the energy efficiency schemes proposed up to now and select one that we believe is more realistic to deploy. Furthermore, we specify the inputs required for the selected “green” routing approach. Moreover, we study existing and new protocols that can provide basic network monitoring functionality that enables the acquirement of these inputs. In the end, a Software Defined Networking (SDN) approach is proposed to facilitate the development of energy-efficient aware networks. The details of a basic SDN monitoring application are presented from an abstract architectural point of view and three designs stemming from this basic architecture are discussed. The three designs are namely All_Flow, First_Switch and Port_FlowRemoved. The first two were implemented as steps towards understanding the full capabilities of performing monitoring in SDN enabled networks and provided useful input towards realizing the third one as a proof of concept. Their usage and faults are discussed as they can provide useful insight for possible future implementations. The Port_FlowRemoved is the design and implementation that is suggested as providing the most fitting results for the monitoring purpose at hand. This purpose is to retrieve the identified inputs for the selected “green” networking approach. The differentiation factor among the three designs is how they collect the required inputs from the network. A fast-prototype is created as a proof of concept in order to validate the proposed architecture and thus empower the validity of the idea.

Several heterogeneous, intelligent and distributed devices can be connected to interact with one another over the internet in what is known as the internet of things (IoT). Also, the concept of IoT can be exploited in the industrial environment for increasing the production output of goods and services and for mitigating the risk of disaster occurrences. This application of IoT for enhancing industrial production is known as industrial IoT (IIoT). More so, the benefits of IoT technology can be particularly exploited across the maritime industry in what is termed the internet of maritime things (IoMT) where sensors and actuator devices are implanted on marine equipment in order to foster the communication efficacy of nautical radio networks. Marine explorations may suffer from unwanted situations such as transactional delays, environmental degradation, insecurity, seaport congestions, accidents and collisions etc, which could arise from severe environmental conditions. As a result, there is a need to develop proper communication techniques that will improve the overall quality of service (QoS) and quality of experience (QoE) of marine users. To address these, the merits of contemporaneous technologies such as ubiquitous computing, software-defined networking (SDN) and network functions virtualization (NFV) in addition to salubrious communication techniques including emergent configurations (EC), channel estimation (CE) and communication routing protocols etc, can be utilized for sustaining optimal operation of pelagic networks. Emergent configuration (EC) is a technology that can be adapted into maritime radio networks to support the operation and collaboration of IoT connected devices in order to improve the efficiency of the connected IoT systems for maximum user satisfaction. To meet user goals, the connected devices are required to cooperate with one another in an adaptive, interoperable, and homogeneous manner. In this thesis, a survey on the concept of IoT is presented in addition to a review of IIoT systems. The applications of ubiquitous computing and SDN technology are employed to design a newfangled network architecture which is specifically propounded for enhancing the throughput of oil and gas production in the maritime ecosystem. The components of this architecture work in collaboration with one another by attempting to manage and control the exploration process of deep ocean activities especially during emergencies involving anthropogenic oil and gas spillages. Several heterogeneous, intelligent and distributed devices can be connected to interact with one another over the internet in what is known as the internet of things (IoT). Also, the concept of IoT can be exploited in the industrial environment for increasing the production output of goods and services and for mitigating the risk of disaster occurrences. This application of IoT for enhancing industrial production is known as industrial IoT (IIoT). More so, the benefits of IoT technology can be particularly exploited across the maritime industry in what is termed the internet of maritime things (IoMT) where sensors and actuator devices are implanted on marine equipment in order to foster the communication efficacy of nautical radio networks. Marine explorations may suffer from unwanted situations such as transactional delays, environmental degradation, insecurity, seaport congestions, accidents and collisions etc, which could arise from severe environmental conditions. As a result, there is a need to develop proper communication techniques that will improve the overall quality of service (QoS) and quality of experience (QoE) of marine users. To address these, the merits of contemporaneous technologies such as ubiquitous computing, software-defined networking (SDN) and network functions virtualization (NFV) in addition to salubrious communication techniques including emergent configurations (EC), channel estimation (CE) and communication routing protocols etc, can be utilized for sustaining optimal operation of pelagic networks. Emergent configuration (EC) is a technology that can be adapted into maritime radio networks to support the operation and collaboration of IoT connected devices in order to improve the efficiency of the connected IoT systems for maximum user satisfaction. To meet user goals, the connected devices are required to cooperate with one another in an adaptive, interoperable, and homogeneous manner. In this thesis, a survey on the concept of IoT is presented in addition to a review of IIoT systems. The applications of ubiquitous computing and SDN technology are employed to design a newfangled network architecture which is specifically propounded for enhancing the throughput of oil and gas production in the maritime ecosystem. The components of this architecture work in collaboration with one another by attempting to manage and control the exploration process of deep ocean activities especially during emergencies involving anthropogenic oil and gas spillages. On the other hand, CE is a utilitarian communication technique that can be exploited during maritime exploration processes which offer additional reinforcement to the capacities of the nautical radio network. This technique enables the receivers of deep-sea networks to efficiently approximate the channel impulse response (CIR) of the wireless communication channel so that the effects of the communication channel on the transmitting aggregated cluster head information can be proficiently understood and predicted for useful decision-making procedures. Two CE schemes named inter-symbol interference/ average noise reduction (ISI/ANR) and reweighted error-reducing (RER) are designed in this study for estimating maritime channels for supporting the communication performances of nautical radio networks in both severe and light-fading environmental conditions. In the proposed RER method, the Manhattan distance of the CIR of an orthodox adaptive estimator is taken, which is subsequently normalised by a stability constant ɛ whose responsibility is for correcting any potential numerical system instability that may arise during the updating stages of the estimation process. To decrease the received signal error, a log-sum penalty function is eventually multiplied by an adjustable leakage (ɛ ) ̈that provides additional stability to the oscillating channel behaviour. The performance of the proposed RER method is further strengthened and made resilient against channel effects by the introduction of a reweighting attractor that further contracts the mean square error of this proposed estimator. In the ISI/ANR technique, the effects of possible ISI that may arise from maritime transmissions is considered and transformed using a low-pass filter that is incorporated for eliminating the effects of channel noise possible effects of multipath propagation. The RER scheme offered superior CE performances in comparison to other customary techniques such as the adaptive recursive least squares and normalised least mean square method in addition to conventional linear approaches such as least squares, linear minimum mean square error and maximum-likelihood estimation method. The proposed ISI/ANR technique offered an improved MSE performance in comparison to all considered linear methods. Finally, from this study, we were able to establish that accurate CE methods can improve the QoS and QoE of nautical radio networks in terms of network data rate and system outage probability. / Thesis (PhD (Computer Engineering))--University of Pretoria, 2021. / University of Pretoria Doctoral research grant, South African National Research Foundation/Research and Innovation Support and Advancement (NRF/RISA) research grant. Center for Connected Intelligence, Advanced Sensor Networks research group, University of Pretoria. / Electrical, Electronic and Computer Engineering / PhD (Computer Engineering) / Unrestricted

In recent years, a new generation of architecture has emerged in the world of computer networks, known as software-defined networking (SDN), that aims to improve and remove the limitations of traditional networks. Although SDN provides viable benefits, it has faced many security threats and vulnerability-related issues. To solve security issues in the SDN, one of the most vital solutions is employing an intrusion detection system (IDS). Merging IDS into the SDN network remains efficient due to the unique features of SDN, such as high manageability, flexibility, and programmability. In this paper, we propose a new approach as a kangaroo-based intrusion detection system (KIDS), which is an SDN-based architecture for attack detection and malicious behaviors in the data plane. Designing a zone-based architecture in the KIDS assists us in achieving a distributed architecture which is scalable in both area and anomaly detection. In the KIDS architecture, the IDS module supplies the flow-based and packet-based intrusion detection components based on monitoring packet parser and Flow tables of the SDN switches. In the proposed approach, the IDS uses consecutive jumps like a kangaroo for announcing the attacks both to the SDN controller and other IDSs, contributing to improved scalability and efficiency. The evaluation of the proposed approach shows an enhanced performance against that of peer approaches in detecting malicious packets.