Дисертації з теми "Service Function Chain Placement"

Recent advancements in mobile communication technology have led to the fifth generation of mobile cellular networks (5G), driven by the proliferation in data traffic demand, stringent latency requirements, and the desire for a fully connected world. This transformation calls for novel technology solutions such as Multi-access Edge Computing (MEC) and Network Function Virtualization (NFV) to satisfy service requirements while providing dynamic and instant service deployment. MEC and NFV are two principal and complementary enablers for 5G networks whose co-existence can lead to numerous benefits. Despite the numerous advantages MEC offers, physical resources at the edge are extremely scarce and require efficient utilization. In this doctoral dissertation, we first attempt to optimize resource utilization at the network edge for the scenario of live video streaming. We specifically utilize the real-time Radio Access Network (RAN) information available at the MEC servers to develop a machine learning-based prediction solution and anticipate user requests. Consequently, Integer Linear Programming (ILP) models are used to prefetch/cache video contents from a centralized video server. Regarding the advantages of NFV technology for the deployment of NFs, the second problem that this dissertation address is the proper association of the users to the gNBs along with efficient placement of SFCs on the substrate network. Our primary purpose is to find a proper embedding of the SFC in a hierarchical 5G network. The problem is formulated as a Mixed Integer Linear Programming (MILP) model, having the objective to minimize service provisioning cost, link utilization, and the effect of VNF migration on users' perceived quality of experience. After rigorously analyzing the proposed SFC placement and considering mobile networks' dynamicity, our next goal is to develop an ILP-based model that minimizes the resource provisioning cost by dynamically embed and scale SFCs so that provisioning cost is minimized while user requirements are met.

5G and beyond mobile network technology promises to deliver unprecedented ultra-low latency and high data rates, paving the way for many novel applications and services. Network Function Virtualization (NFV) and Multi-access Edge Computing (MEC) are two technologies expected to play a vital role in achieving ambitious Quality of Service requirements of such applications. While NFV provides flexibility by enabling network functions to be dynamically deployed and inter-connected to realize Service Function Chains (SFC), MEC brings the computing capability to the mobile network's edges, thus reducing latency and alleviating the transport network load. However, adequate mechanisms are needed to meet the dynamically changing network service demands (i.e., in single and multiple domains) and optimally utilize the network resources while ensuring that the end-to-end latency requirement of services is always satisfied. In this dissertation work, we break the problem into three separate stages and present the solutions for each one of them.Firstly, we apply Artificial Intelligence (AI) techniques to drive NFV resource orchestration in MEC-enabled 5G architectures for single and multi-domain scenarios. We propose three deep learning approaches to perform horizontal and vertical Virtual Network Function (VNF) auto-scaling: (i) Multilayer Perceptron (MLP) classification and regression (single-domain), (ii) Centralized Artificial Neural Network (ANN), centralized Long-Short Term Memory (LSTM) and centralized Convolutional Neural Network-LSTM (CNN-LSTM) (single-domain), and (iii) Federated ANN, federated LSTM and federated CNN-LSTM (multi-domain). We evaluate the performance of each of these deep learning models trained over a commercial network operator dataset and investigate the pros and cons of different approaches for VNF auto-scaling. For the first approach, our results show that both MLP classifier and MLP regressor models have strong predicting capability for auto-scaling. However, MLP regressor outperforms MLP classifier in terms of accuracy. For the second approach (one-step prediction), CNN-LSTM performs the best for the QoS-prioritized objective and LSTM performs the best for the cost-prioritized objective. For the second approach (multi-step prediction), the encoder-decoder CNN-LSTM model outperforms the encoder-decoder LSTM model for both QoS and Cost prioritized objectives. For the third approach, both federated LSTM and federated CNN-LSTM models perform equally better than the federated ANN model. It was also noted that in general federated learning approaches performs poorly compared to centralized learning approaches. Secondly, we employ Integer Linear Programming (ILP) techniques to formulate and solve a joint user association and SFC placement problem, where each SFC represents a service requested by a user with end-to-end latency and data rate requirements. We also develop a comprehensive end-to-end latency model considering radio delay, backhaul network delay and SFC processing delay for 5G mobile networks. We evaluated the proposed model using simulations based on real-operator network topology and real-world latency values. Our results show that the average end-to-end latency reduces significantly when SFCs are placed at the ME hosts according to their latency and data rate demands. Furthermore, we propose an heuristic algorithm to address the issue of scalability in ILP, that can solve the above association/mapping problem in seconds rather than hours.Finally, we introduce lightMEC - a lightweight MEC platform for deploying mobile edge computing functionalities which allows hosting of low-latency and bandwidth-intensive applications at the network edge. Measurements conducted over a real-life test demonstrated that lightMEC could actually support practical MEC applications without requiring any change to existing mobile network nodes' functionality in the access and core network segments. The significant benefits of adopting the proposed architecture are analyzed based on a proof-of-concept demonstration of the content caching use case. Furthermore, we introduce the AI-driven Kubernetes orchestration prototype that we implemented by leveraging the lightMEC platform and assess the performance of the proposed deep learning models (from stage 1) in an experimental setup. The prototype evaluations confirm the simulation results achieved in stage 1 of the thesis.

L'émergence de la technologie de virtualisation des fonctions réseau (NFV) a suscité un vif intérêt autour de la conception, la gestion et le déploiement de services réseau de manière flexible, automatisée et indépendante du fournisseur. La mise en œuvre de la technologie NFV devrait être une solution profitable pour les fournisseurs de services et les clients. Cependant, ce changement de paradigme, amorcé par NFV, nécessite un abandon progressif des services réseau fournis à travers des équipements dédiés. En contrepartie, un environnement totalement ou partiellement virtualisé est proposé pour instancier dynamiquement et à la demande des modules logiciels appelés fonctions de réseau virtuelles (VNF). Cette évolution soulève un ensemble de défis liés au déploiement et à l'exploitation de services, tels que l'orchestration et la gestion, la résilience des services, le contrôle de la qualité de service (QoS), l’approvisionnement des ressources, etc. En outre, la question centrale à résoudre dans le contexte NFV est la suivante : « comment placer et chaîner effacement des fonctions virtuelles d’un service afin de fournir un niveau de qualité demandé par le client tout en optimisant l'utilisation des ressources par le fournisseur de services ? ”.Ainsi, cette thèse étudie la problématique du placement et du chaînage des VNF en tenant compte de certaines exigences de service telles que le délai de bout en bout, la disponibilité du service et la consommation d'énergie, et propose un ensemble d'algorithmes et de mécanismes visant à optimiser le déploiement des services demandés/fournis. Nos contributions dans cette thèse sont triples. Premièrement, nous proposons deux algorithmes de placement et de chaînage de VNF sensibles au délai de bout-en-bout pour des applications temps-réel. Les algorithmes proposés visent à respecter le délai approprié de bout-en-bout qui dépend du service déployé (exemples : VoIP, Streaming, etc.). Deuxièmement, nous présentons une analyse comparative de la disponibilité des services et nous proposons deux mécanismes de placement et de chaînage de VNF pour garantir un niveau prédéfini de disponibilité. L’objectif est de fournir des services résilients en ajustant avec précision les paramètres du schéma de protection (nombre, type, emplacement et taille des instances VNF) nécessaires pour atteindre ce niveau de disponibilité en dépit des défaillances du réseau. Enfin, nous proposons une architecture générale qui explore la possibilité d’étendre le paradigme de la virtualisation à l’Internet des objets (IoT). À cette fin, nous définissons un mécanisme de placement et de chaînage respectant les contraintes énergétiques pour des services IoT. Notre architecture propose de découpler et de virtualiser les fonctionnalités inhérentes à un objet connecté de l’équipement IoT physique. En étendant NFV au domaine IoT, notre solution ouvre de nouvelles perspectives d’application en supportant de nouveaux cas d’usages
The emergence of Network Function Virtualization (NFV) technology has aroused keen interest to design, manage and deploy network services in a flexible, automated and vendor-agnostic manner. Implementing NFV technology is expected to be a win-win solution for both service providers and costumers. However, this paradigm shift, sparked by NFV, calls for a progressive abandon of network services that are provided as hardware appliance and rather it proposes a fully or partially virtualized environment that offers software modules called Virtual Network Functions (VNFs). This shift rises a set of challenges related to service deployment and operation such as orchestration and management, service resiliency, Quality of Service (QoS) and resource provisioning among others. Furthermore, the core question that needs to be solved within NFV context is “What is the best way to place and chain VNFs that form a service in order to meet Service Level Agreement requirements (costumer side) while optimizing resource usage (service provider side)?”.This thesis investigates the problem of VNF Placement and Chaining considering service requirements such as end-to-end delay, service availability and energy consumption and proposes a set of algorithms and mechanisms that aim to achieve an optimized deployment of the requested/provided services. Our contributions in this thesis are threefold. First, we propose a delay-aware Placement and Chaining algorithms for delay-sensitive applications over NFV networks. The proposed algorithms aim to meet the appropriate end-to-end delay defined according to the deployed service (VoIP, Streaming, etc.). Second, we provide a comprehensive service availability benchmarking and we propose two availability-aware mechanisms for VNFs chain. The aim is to provide resilient service provisioning by fine-tuning the parameters of the protection scheme (the number, the type, the placement and the size of the spare instances) needed to reach a predefined availability level, despite network failures. Finally, we propose a framework architecture that explores the possibility to extend the virtualization paradigm to Internet of Things (IoT). Toward this end, we define an energy-aware Placement and Chaining for IoT services where inherent IoT functionalities are decoupled from specific dedicated IoT devices and instantiated on-demand. By bringing together NFV and IoT paradigms, this extension opens new perspectives and push toward designing new use cases

A Virtualização de Funções de Rede (NFV – Network Function Virtualization) é um novo conceito arquitetural que está remodelando a operação de funções de rede (e.g., firewall, gateways e proxies). O conceito principal de NFV consiste em desacoplar a lógica de funções de rede dos dispositivos de hardware especializados e, desta forma, permite a execução de imagens de software sobre hardware de prateleira (COTS – Commercial Off-The-Shelf). NFV tem o potencial para tornar a operação das funções de rede mais flexíveis e econômicas, primordiais em ambientes onde o número de funções implantadas pode chegar facilmente à ordem de centenas. Apesar da intensa atividade de pesquisa na área, o problema de posicionar e encadear funções de rede virtuais (VNF – Virtual Network Functions) de maneira escalável e com baixo custo ainda apresenta uma série de limitações. Mais especificamente, as estratégias existentes na literatura negligenciam o aspecto de encadeamento de VNFs (i.e., objetivam sobretudo o posicionamento), não escalam para o tamanho das infraestruturas NFV (i.e., milhares de nós com capacidade de computação) e, por último, baseiam a qualidade das soluções obtidas em custos operacionais não representativos. Nesta tese, aborda-se o posicionamento e o encadeamento de funções de rede virtualizadas (VNFPC – Virtual Network Function Placement and Chaining) como um problema de otimização no contexto intra- e inter-datacenter. Primeiro, formaliza-se o problema VNFPC e propõe-se um modelo de Programação Linear Inteira (ILP) para resolvêlo. O objetivo consiste em minimizar a alocação de recursos, ao mesmo tempo que atende aos requisitos e restrições de fluxo de rede. Segundo, aborda-se a escalabilidade do problema VNFPC para resolver grandes instâncias do problema (i.e., milhares de nós NFV). Propõe-se um um algoritmo heurístico baseado em fix-and-optimize que incorpora a meta-heurística Variable Neighborhood Search (VNS) para explorar eficientemente o espaço de solução do problema VNFPC. Terceiro, avalia-se as limitações de desempenho e os custos operacionais de estratégias típicas de aprovisionamento ambientes reais de NFV. Com base nos resultados empíricos coletados, propõe-se um modelo analítico que estima com alta precisão os custos operacionais para requisitos de VNFs arbitrários. Quarto, desenvolve-se um mecanismo para a implantação de encadeamentos de VNFs no contexto intra-datacenter. O algoritmo proposto (OCM – Operational Cost Minimization) baseia-se em uma extensão da redução bem conhecida do problema de emparelhamento ponderado (i.e., weighted perfect matching problem) para o problema de fluxo de custo mínimo (i.e., min-cost flow problem) e considera o desempenho das VNFs (e.g., requisitos de CPU), bem como os custos operacionais estimados. Os resultados alcaçados mostram que o modelo ILP proposto para o problema VNFPC reduz em até 25% nos atrasos fim-a-fim (em comparação com os encadeamentos observados nas infra-estruturas tradicionais) com um excesso de provisionamento de recursos aceitável – limitado a 4%. Além disso, os resultados evidenciam que a heurística proposta (baseada em fix-and-optimize) é capaz de encontrar soluções factíveis de alta qualidade de forma eficiente, mesmo em cenários com milhares de VNFs. Além disso, provê-se um melhor entendimento sobre as métricas de desempenho de rede (e.g., vazão, consumo de CPU e capacidade de processamento de pacotes) para as estratégias típicas de implantação de VNFs adotadas infraestruturas NFV. Por último, o algoritmo proposto no contexto intra-datacenter (i.e. OCM) reduz significativamente os custos operacionais quando comparado aos mecanismos de posicionamento típicos uti
Network Function Virtualization (NFV) is a novel concept that is reshaping the middlebox arena, shifting network functions (e.g. firewall, gateways, proxies) from specialized hardware appliances to software images running on commodity hardware. This concept has potential to make network function provision and operation more flexible and cost-effective, paramount in a world where deployed middleboxes may easily reach the order of hundreds. Despite recent research activity in the field, little has been done towards scalable and cost-efficient placement & chaining of virtual network functions (VNFs) – a key feature for the effective success of NFV. More specifically, existing strategies have neglected the chaining aspect of NFV (focusing on efficient placement only), failed to scale to hundreds of network functions and relied on unrealistic operational costs. In this thesis, we approach VNF placement and chaining as an optimization problem in the context of Inter- and Intra-datacenter. First, we formalize the Virtual Network Function Placement and Chaining (VNFPC) problem and propose an Integer Linear Programming (ILP) model to solve it. The goal is to minimize required resource allocation, while meeting network flow requirements and constraints. Then, we address scalability of VNFPC problem to solve large instances (i.e., thousands of NFV nodes) by proposing a fixand- optimize-based heuristic algorithm for tackling it. Our algorithm incorporates a Variable Neighborhood Search (VNS) meta-heuristic, for efficiently exploring the placement and chaining solution space. Further, we assess the performance limitations of typical NFV-based deployments and the incurred operational costs of commodity servers and propose an analytical model that accurately predict the operational costs for arbitrary service chain requirements. Then, we develop a general service chain intra-datacenter deployment mechanism (named OCM – Operational Cost Minimization) that considers both the actual performance of the service chains (e.g., CPU requirements) as well as the operational incurred cost. Our novel algorithm is based on an extension of the well-known reduction from weighted matching to min-cost flow problem. Finally, we tackle the problem of monitoring service chains in NFV-based environments. For that, we introduce the DNM (Distributed Network Monitoring) problem and propose an optimization model to solve it. DNM allows service chain segments to be independently monitored, which allows specialized network monitoring requirements to be met in a efficient and coordinated way. Results show that the proposed ILP model for the VNFPC problem leads to a reduction of up to 25% in end-to-end delays (in comparison to chainings observed in traditional infrastructures) and an acceptable resource over-provisioning limited to 4%. Also, we provide strong evidences that our fix-and-optimize based heuristic is able to find feasible, high-quality solutions efficiently, even in scenarios scaling to thousands of VNFs. Further, we provide indepth insights on network performance metrics (such as throughput, CPU utilization and packet processing) and its current limitations while considering typical deployment strategies. Our OCM algorithm reduces significantly operational costs when compared to the de-facto standard placement mechanisms used in Cloud systems. Last, our DNM model allows finer grained network monitoring with limited overheads. By coordinating the placement of monitoring sinks and the forwarding of network monitoring traffic, DNM can reduce the number of monitoring sinks and the network resource consumption (54% lower than a traditional method).

Thesis (MBA (Business Management))--Stellenbosch University, 2009.
ENGLISH ABSTRACT: A typical medium-sized service company in South Africa faces definite challenges and misses opportunities in optimising and managing its operational expenses in its sourcing and supply chain. Medium-sized companies generally do not have the benefit of sophisticated in-house procurement divisions and extensive buying power but they still have to optimise their expenses in order to ensure maximum profitability and business longevity. The study addresses these challenges and opportunities by developing and proposing a framework for a typical medium-sized service company that does not have a central procurement function related to the cost optimisation and management of the identified sourcing and supply-chain elements. The research process entails the development of a conceptual framework developed from literature analysis and a representational practical environment as well as the testing, reviewing and validating of the test implementation of the framework. The results flowing from the testing and implementation of the framework are excluded from the research project. The results are displayed in the form of a summary of specific relevant information around each framework element as well as a conceptual framework. Recommendations are made in relation to the implementation and practical adaptation of the framework. It is concluded that a medium-sized service company can successfully optimise and manage its costs in the sourcing and supply-chain elements in the absence of a central procurement function. This requires the effective implementation of the developed framework, which was adapted from sophisticated larger companies. Implementation and the measurement process need to be driven or, preferably, conducted by the medium-sized company owner or management. Buy-in of the framework and the processes around it needs to be obtained from employees, suppliers and other stakeholders.
AFRIKAANSE OPSOMMING: ’n Tipiese mediumgrootte diensverskaffingsonderneming in Suid-Afrika staar spesifieke uitdagings in die gesig en ontbeer geleenthede met betrekking tot die optimalisering en bestuur van die bedryfsuitgawes in sy aankoop- en verskaffingsketting. Mediumgrootte ondernemings beskik in die algemeen nie oor die voordeel van ’n gesofistikeerde interne aankoopafdeling en groot aankoopkrag nie, maar hierdie ondernemings moet nog steeds hulle koste-uitgawes optimaliseer in die nastreef van maksimale wins en besigheidsukses. Die navorsingstudie neem hierdie uitdagings en geleenthede onder die loep deur die ontwikkeling en aanbeveling van ’n raamwerk vir ’n tipiese mediumgrootte diensverskaffingsonderneming wat nie oor ’n sentrale interne aankoopafdeling beskik nie. Die raamwerk is gebaseer op koste-optimalisering en -bestuur van die geïdentifiseerde aankoop- en verskaffingskettingelemente. Die navorsingsproses behels die ontwikkeling van ’n konseptuele raamwerk deur literatuuranalise en ’n verteenwoordigende praktiese omgewing, sowel as die toetsing, hersiening en geldigverklaring van die raamwerk tydens die toets-implementering daarvan in die praktiese omgewing. Die resultate van die toets en implementering van die raamwerk vorm nie deel van hierdie studie nie. Die navorsingsresultate word ten toon gestel in die vorm van ’n opsomming van die relevante raamwerkelemente sowel as ’n ontwikkelde raamwerk. Aanbevelings ten opsigte van die implementering en praktiese aanpassing van die raamwerk word dan gemaak. Die gevolgtrekking word gemaak dat ’n mediumgrootte diensverskaffingsonderneming wel in die afwesigheid van ’n sentrale interne aankoopafdeling sy kostes in die aankoop- en verskaffingsketting suksesvol kan optimaliseer en bestuur. Dit kan gedoen word deur die doeltreffende implementering en aanwending van die ontwikkelde raamwerk wat vanaf groter ondernemings aangepas is. Die raamwerk se implementering- en metingsproses moet deur die eienaar of bestuur van die mediumgrootte onderneming gelei, of verkieslik uitgevoer, word. Die werknemers, verskaffers en ander belanghebbendes se ondersteuning vir die raamwerk en gepaardgaande prosesse is baie belangrik vir die implementering en sukses van die raamwerk.

碩士
國立臺灣科技大學
資訊工程系
106
Software-defined network(SDN) is shifting data communication in network toward more centralized control. The controller manages all necessary information and decides the transit path through commands. Middleboxes such as firewalls, decryptors, load balancers, etc., are functional hardware devices that are　indispensable in the cloud　and in corporate networks, but are expensive and difficult to maintain. The availability of virtual machines enables dynamic placement of virtual network functions(VNFs) on demand, it can reduce a large number of manual configuration processes, introduce flexibility, and increase deployment efficiency. Service requirements may need more than one VNF. Instead, it may need to traverse a set of sequential functional services called Service Function Chain (SFC). For example: first through the firewall, then through the encryption device, and finally through the decryption device, if these functional services are unordered, it will affect　the correctness of the entire service requirements. A key challenge is to efficiently route service function chains and place processing functions in a network under operational constraints. We must overcome two challenges: (1) determining a flow path that traverses suitable processing nodes in the correct order to meet the requirements of a given service chain, and (2) considering network load and other dynamic characteristics when routing through 　existing VNFs. We present a method to solve the node-constrained service chain routing problem in a software-defined network. The main idea is to transform the network representation to a layered graph that considers processing steps and allows the use of conventional shortest path algorithms, such as Dijkstra’s algorithm, to solve the problem. We are not only pursuing the high efficiency of acceptance ratio, but also 　taking into account the execution time of the algorithm.

碩士
國立清華大學
資訊工程學系
104
Network function virtualization (NFV) has changed the traditional usage of network functions from hardware appliances to software implementation. Placement of virtual network functions (VNFs) has become a fundamental problem in NFV. In this study, relation in VNFs, including dependency and changes of flow size, would be important factors in placement of VNFs onto virtual machines on servers. Metric embedding is adopted to combine the relation of VNFs and virtual machines on network servers. The structures shown in service chains are seized by embedding algorithm and thus reflect the result on placement of VNFs on the network. In the simulation section, the proposed technique is evaluated and performs well in test cases, showing that the proposed technique gives a fine solution to VNF placement problem.

碩士
國立交通大學
網路工程研究所
104
Network function virtualization (NFV) has drawn much attention in recent years; some network functions used to be deployed on the specific hardware have become virtualized instances on general servers to achieve more scalability and flexibility. In data center, service function chaining (SFC) makes different workflow traverse different network functions in a specific ordering to provide different levels of the service for its customer. Since the distance between any adjacent network functions in a service chain will decide the total bandwidth consumption for that chain, the placement of the virtualized network functions in data center becomes a major issue. In this thesis, we treat this placement problem as multi-layer bin packing problem and formulate it as Integer Linear Programming. We propose three algorithms for different network topologies. Partition Best-fit is designed for the topology-unaware, and Multi-layer Best-fit and Multi-layer Worst-fit are designed for the tree-like network topology. The experiment results show that both Partition Best-fit and Multi-layer Best-fit can reduce bandwidth consumption in 15% while only increase a little number of used servers in 1% compare to traditional Best-fit algorithm.

碩士
國立中央大學
資訊工程學系
105
In recent years, the network functions virtualization (NFV) was proposed to separate network functions from the underlying hardware to provide value-added service functions with software. The concept of service function chain (SFC) architecture has also been proposed, which allow traffic pass through some value-added network functions between end-to-end services. The placement of the service functions in data centers significantly affects the performance of SFC. We study the service function placement problem in hybrid NFV environment, which is an alternative plan during the transition from traditional networks to NFV networks in data centers. We present GASFP, a high-performance approach for optimizing service function placement and service chaining in hybrid NFV environment. GASFP combine GA (Genetic Algorithm) and Greedy best-first search algorithm to reduce the waste of data center computation resource for providing SFC services and find a good solution to slow down the growth of delay from unsuitable service function placement. Simulation results show that when compared with other 3 algorithms: BINPacking, GKMSFP and VCD, GASFP has the best performance of average delay and VNF placement cost no matter in hybrid NFV environment or pure NFV environment. We also implement a service function cloud platform and a SFC steering mechanism named NSHSFC in XenServer to verify our works are more comprehensive and realistic than the former works. Keyword: Software-defined Networking; Service Function Chain; Network Function Virtualization; service function placement; genetic algorithm

碩士
國立中央大學
資訊工程學系
103
In recent years, as growth of the cloud computing technology, more demand for network resources allocation and management of value-added services are appeared. The new concept of network service architecture, Service Function Chain (SFC), has been proposed. The specification of various vendor equipments is unique, prompts the traditional network to change. The purpose of Software-Defined Network (SDN) is to change the physical, complex network into a virtual, programmable and open network architecture. OpenFlow protocol is the most common technology for SDN architecture. The Network management can be more flexible by the centralized and programable property of OpenFlow. VLAN and OpenFlow based Service Function Chain (VOFSFC) and services deployment mechanism are proposed to improve the flexibility of the network service deployment and reduce the consumption of network resources. VOFSFC use the virtual LAN (VLAN) technology to develop VLAN of Path adoption mechanism. It can identify different requirements of network traffic, steer/deliver packets to service functions and record information of the traffic. In addition, consider the placement of service functions, develop and propose GARCH and K-Means Service Function Placement (GKMSFP) mechanism to cluster and deploy service functions for reduce the load of network resources. The results shows the proposed VOPSFC system can provide high elasticity and flexibility for the network management, and the proposed GKMSFP mechanism can reduce 45.6% network resource consumption from the general situation which without cluster mechanism. It shows that mechanisms can reduce the traffic load of network in the SFC system.

碩士
國立中正大學
資訊工程研究所
107
Network function virtualization (NFV) is a new architecture that implements network functions in a virtialized environment. However, the platform for deploying network functions may be insecure. To eliminate the threat, network functions can be deployed with a trusted execution environment (TEE) mechanism such as Intel Software Guard Extension (SGX) in a secure region called enclave. In existing studies, all the network functions in a service chain are deployed in one enclave, but the enclave size is limited (up to 128MB in our investigation). In addition, the network functions need to be re-implemented from scratch to meet the requirements. In this work, we propose an approach, a service function chain with SGX-based containers, and study the performance issues on it. We deploy the network functions in several situations: on the same host, in the same container or on different hosts. We estimate the performance of packet transmission between the network functions in the same container with the shared memory mechanism (emulated by RAMdisk). In the other situations, the packets are transmitted via a switch and/or the virtual network interface controller (NIC). We also build two open-source network functions in SGX-based containers with few modifications for demonstration. In the experiments, we evaluate our system and find the throughput in shared memory can be as high as 400MB/s in packet transmission. We also find the encryption operation (AES in our experiments) and the receiving buffer size also have an impact on the throughput.

碩士
國立宜蘭大學
資訊工程學系碩士班
105
In recent years, mobile communication technology continues to explode the progress of the relative, the network management costs are growing, because the network needs to configure a lot of settings, such as: Route, VLAN (Virtual Local Area Network), QoS (Quality Of Service), STP (Spanning Tree Protocol) and LACP (Link Aggregation Control Protocol), etc., in a large network, because the traditional network of each node settings are rigid configuration, there is no way to achieve the global vision , Resulting in network administrators from time to time need to change these configurations, for example, whenever the New Year, the crowd brought huge amounts of information poured into, causing some switches overload and collapse. SDN (Software-defined networking) is a new concept of network architecture, the main purpose is to network Flexibility, Manageability, Scalability, SDN is the biggest feature of the traditional network is divided into the control layer (Control Plane) and the data layer (Data Plane), the data forwarding decision by a single or multiple Controller to deal with, as long as the control layer and the data layer through the same open protocol OpenFlow communication, it is no longer rigidly adhere to the hardware label or model, the administrator or programmer can write software to easily achieve load balancing, so the appearance of the SDN improved Traditional networks can only be partially optimized and can not be fully configured shortcomings. Considering the network packet of the various forwarding are bundled behind a number of network services, so we think that the use of SDN project Service Function Chain architecture so that the data can be flexible to do the forwarding action, select Service Function Chain is different from the use of flow table , the flow can also be real-time conversion when the traffic generated , even when the user made a different request, but also Automatically redirect packets to the desired service, in order to achieve the balance of traffic mechanism, the use of Network Service Header attached to the data, so that the data can be indicated by the Network Service Header to go through the node, the experimental structure of this study based on an physical machine simulation of eight virtual machines, through the design of the greedy algorithm based on the service function chain to add and modify the direction and order of the various services forward, after the experimental results, the use of this method can effectively relieve traffic in high-density environment caused by congestion problems.

碩士
國立臺灣科技大學
資訊工程系
106
Due to swift Internet technology and computer hardware development in recent years, the demands of network functions are rising. Compared to hardware-based network services, it is faster and more flexible to virtualize the network services and a flows may be processed by multiple network services as a Service Function Chain (SFC). Using Software Defined Network (SDN) is highly popular in the world because it can reduce deployment cost and time. However, the performance of Network Function Virtualiza- tion will be troubled by some congenital drawbacks. For instance, Network Function Virtualization usually suffers from I/O latency because it causes lots of interrupts that decreases CPU utilization. Although this can be solved somehow such as using SR-I/OV network card to reduce the system interrupts, the methods are still not perfect because it is not compatible with existing Software Defined Network system. For example, we often runs OpenvSwitch, an OpenFlow virtual switch supporting Software Defined Network, in general purpose server’s host OS to manage all the network packets forwarded to Virtualized Network Functions (VNFs). However, with OpenvSwitch, network packets must bypass host OS to guest OS when we use SR-I/OV network card. We will lose the benefit of SR-I/OV to enable SDN by using OpenvSwitch. Therefore, we propose an integrated architecture from a wider perspective in system design to overcome main defects described above. We modify appropriate components in SR-I/OV driver and OpenvSwitch to make Service Function Chain constructed by Virtualized Network Functions perform better and compatible with Software Defined Network environment.

碩士
大葉大學
事業經營研究所碩士在職專班
96
This research as 7-11 as research object, take advantage of Quality Function Deployment systematize inquire into 7-11 quality of service and quality of technology, questionnaire survey object as go pass 7-11 do shop consumer. Research framework refer PZB quality of service's mode, and refer Dabbolkar(1996) bring up de quality of service's measuring watch Retail Service Quality Scale(RSQS) come up to relative to document, again depend on chain stores characteristic, increase or curtail ask, analyze as questionaire project. Finding by SPSS calculate behind, as factor analysis establish 7-11 quality of service evaluate mode can divide into entity's image, credibility, personnel interaction, problem settle, shop's policy, price of article wait for six construct, permeate each quality of service construct element boot Quality Function Deployment, will consumer need de serve quality requirement transform for 7-11 quality's technology request, at last integrate analyze result hear of 7-11 front ten important quality's technologies ameliorate project for: Attitude of service, at scene handling capacity, personnel education and training, check system, master market's ability, manpower preparation, advertisement filtrate ability, enterprise's brand's image, return goods disposal, prompt consuming power. This findings will can supply 7-11 proceed what positional and quality of service administer's referrence.

碩士
輔仁大學
資訊管理學系
101
The purpose of this study is to understand the consumers’ acceptance over self-order service and the key function needs. Hopefully the result of this study would furnish the chain fast food restaurants with a good reference regarding to offering self-service. Based on Technology Acceptance Model (TAM), Direct Benefit and Deferred Benefit are adopted as external variables for Perceived Usefulness and User Interface and Information Content are adopted as external variables for Perceived Ease of Use. Totally 130 responses were collected via an internet based questionnaire. Among these responses, 109 responses are valid. Regression and Partial Least Squares (PLS) were used to analyze the responses collected. According to the responses, Deferred Benefit is positively related to Direct Benefit and Direct Benefit is positively related to Perceived Usefulness. Information Content is positively related to User Interface and User Interface is positively related to Perceived Ease of Use. Comparing the results with TAM, this study found that Intention to Use is more significantly positively related to Perceived Ease of Use than to Perceived Usefulness. This study also found that Direct Benefit, such as discount, and User Interface, such as provide content for the latest order, are very important for consumers. The results provide the chain fast food restaurants a useful reference for providing self-service.

碩士
國立中央大學
資訊工程學系
105
近年來雲端計算的技術興起，越來越多的服務功能被開發出來，因此有了新的網路服務技術出現，服務功能鏈（Service Function Chain），有了這個技術，就可以佈署更加彈性的服務功能，不必限制於實體場景。軟體定義網路（Software Defined Network），為集中式的網路架構，目的將網路管理變得更加彈性且可程式化，減輕網路管理員的負擔。網路功能虛擬化（Network Function Virtualization）把原本需要特定硬體的網路功能分離出來，用軟體的方式實作，網路功能不再受限於硬體設備商。由於如何達成服務功能鏈的方式尚未成為標準，本論文結合軟體定義網路以及網路功能虛擬化這兩個技術實作出服務功能鏈平台。 本論文除了實作出管理平台還提出了資源管理的方式，有負責high availability的share和recovery機制，有負責提升服務功能使用率的merge機制，以及相對應產生的separate機制，還有應付大量流量的auto-scaling機制scale up和scale down機制。並提出一個演算法STSFCM（Stable Time Service Function Chain Management）用來決定上述六項機制要在何時啟動，在auto-scaling的部分加上時間的因素，使得在穩定使用率的狀態可以更好的利用VNF的資源。

碩士
國立臺灣科技大學
管理技術研究所
86
Under the competitive situation on chain convenient stores, understanding and satisfying customer''s requirements is the key point to win the consumption of the market among the enterprises. However, making use of finite resources to obtain the greatest customer satisfaction needs much effort to take research. This research adopts quality function deployment (QFD) to help the enterprises find the most effective service quality improvement strategy and provide an objective method to evaluate. It makes customer requirements and quality improvement skills combined more rigorously. It also consults the bright and dull competitive situation to improve the effect and the practicality of the service quality improvement strategy. Through the intermediary of quality deployment, skill deployment, service orientation, strategy arrangement and house of quality (HOQ) which have systematization implementation procedures to present the evaluation of the process and the result, this research provides the proprietors an improvement tool which is easy to use and effective to substitute subjective consciousness evaluation method formerly. Hence, it can meet customer''s requirements more quickly and effectively. According to QFD changes, 21 items on the service quality of the chain convenient store transfers to 32 items of the quality improvement skill; this research generates results from the evaluation of HOQ by choosing the following priority of service quality improvement strategy: competition strategy, self-diagnose and achievement evaluation, point of sale system design and application, franchise chain management, race encouragement of track record, voluntary chain management, and strategy management of running stores.

While decision-making under uncertainty is a major universal problem, its implications in the field of transportation systems are especially enormous; where the benefits of right decisions are tremendous, the consequences of wrong ones are potentially disastrous. In the realm of highway systems, decisions related to the highway configuration (number of lanes, right of way, etc.) need to incorporate both the traffic demand and land price uncertainties. In the literature, these uncertainties have generally been modeled using the Geometric Brownian Motion (GBM) process, which has been used extensively in modeling many other real life phenomena. But few scholars, including those who used the GBM in highway configuration decisions, have offered any rigorous justification for the use of this model. This thesis attempts to offer a detailed analysis of various aspects of transportation systems in relation to decision-making. It reveals some general insights as well as a new concept that extends the notion of opportunity cost to situations where wrong decisions could be made. Claiming deficiency of the GBM model, it also introduces a new formulation that utilizes a large and flexible parametric family of jump models (i.e., Lévy processes). To validate this claim, data related to traffic demand and land prices were collected and analyzed to reveal that their distributions, heavy-tailed and asymmetric, do not match well with the GBM model. As a remedy, this research used the Merton, Kou, and negative inverse Gaussian Lévy processes as possible alternatives. Though the results show indifference in relation to final decisions among the models, mathematically, they improve the precision of uncertainty models and the decision-making process. This furthers the quest for optimality in highway projects and beyond.