Large capacity high voltage power transformers are one of the most expensive items of equipment in an electrical power network. Power utilities can ill-afford breakdown of transformers, especially, in a deregulated scenario. The consequences of such a failure are well known. Under these circumstances, utilities have figured-out that condition-based monitoring and diagnosis is worth pursuing, in spite of increased expenditure. Thus, monitoring and diagnosis is an integral part of operation and maintenance. Mechanical forces generated during short-circuits is the main cause leading to displacement/deformation of windings. Frequency response measurements have attained worldwide acceptance as a highly sensitive monitoring tool for detecting occurrence of such events. This is evident from the fact that customized commercial equipment are available (popularly called FRA or SFRA instruments), and with recent introduction of an IEEE draft trial-use guide for application and interpretation of frequency response analysis. Once a damage is detected, the next task is to identify its location along the winding and, if possible, determine its extent of severity. Understandably, these two tasks are best achieved, without disassembling the transformer and should ideally be based on off-line and on-site terminal measurements. In this regard, literature analysis reveals that recent research efforts have successfully demonstrated possibilities of using frequency response data for localization of discrete change in windings. This is indeed noteworthy, in spite of one major drawback. This pertains to excessive computing time needed to synthesize large-sized ladder-network, which automatically limits its practical use. Keeping these issues in mind, a research was initiated to find alternatives. The primary objective of this thesis is to examine the use of- • Complexnetwork-function-lociforlocalizationofadiscretechangeinasingle,isolatedtransformerwinding,basedonterminalmeasurements It goes without saying that the proposed method should be non-invasive, simple, time-efficient and overcome drawbacks in the earlier approach. A brief summary of the proposed method follows- This thesis presents a different approach to tackle the problem of localization of winding deformation in a transformer. Within the context of this thesis, winding deformation means, a discrete and specific change imposed at a particular position on the winding. The proposed method is based on the principle of pre-computing and plotting the complex network-function-loci (e.g. driving-point-impedance) at a selected frequency, for a meaningful range of values for each element (increasing and decreasing) of the ladder network. This loci diagram is called the nomogram. After introducing a discrete change (to simulate a deformation), the driving-point-impedance (amplitude and phase) is measured again .By plotting this single measurement on the nomogram, it is straightforward to estimate the location and identify the extent of change. In contrast to the earlier approach (wherein the entire ladder-network had to be synthesized for every new measurement), the proposed method overcomes the drawbacks, is non-iterative and yields reasonably accurate localization. Experimental results on a model coil and two actual transformer windings (continuous-disc and interleaved-disc) were encouraging and demonstrate its potential. Further details are presented in the thesis.

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

New services and ever increasing traffic volumes require the next generation of mobile networks, e.g. 5G, to be much more flexible and scalable. The primary enabler for its flexibility is transforming network functions from proprietary hardware to software using modern virtualization technologies, paving the way of virtual network functions (VNF). Such VNFs can then be flexibly deployed on cloud data centers while traffic is routed along a chain of VNFs through software-defined networks. However, such flexibility comes with a new challenge of allocating efficient computational resources to each VNF and optimally placing them on a cluster. In this thesis, we argue that, to achieve an autonomous and efficient performance optimization method, a solid understanding of the underlying system, service chains, and upcoming traffic is required. We, therefore, conducted a series of focused studies to address the scalability and performance issues in three stages. We first introduce an automated profiling and benchmarking framework, named NFV-Inspector to measure and collect system KPIs as well as extract various insights from the system. Then, we propose systematic methods and algorithms for performance modelling and resource recommendation of cloud native network functions and evaluate them on a real 5G testbed. Finally, we design and implement a bottom-up performance simulator named PerfSim to approximate the performance of service chains based on the nodes’ performance models and user-defined scenarios. / <p>Article 5 part of thesis as manuscript, now published.</p>

京都大学 / 新制・課程博士 / 博士(情報学) / 甲第24941号 / 情博第852号 / 新制||情||143(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 原田 博司, 教授 佐藤 高史 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

Software Defined Networking (SDN) has become increasingly popular in combination with Network Function Virtualization (NFV). SDN is a way to make a network more programmable and dynamic. However, in order to create a homogeneous network using this concept, legacy equipment will have to be substituted by SDN equipment, which is costly. To close the gap between the legacy world and SDN, we introduce the concept of a legacy Network Management System (NMS) that is connected to an SDN controller to perform service provisioning. This way, the NMS is capable of configuring both legacy as well as SDN networks to provide customers with the services that they have ordered, while still allowing for new SDN features in the SDN domain of the network. The main service we wish to provide using SDN is Service Function Chaining (SFC). Service provisioning consists of dynamically constructing a path through the ordered network services, in this case Virtual Network Functions (VNFs). This thesis focuses on the SDN controller and its interaction with the NMS. This project aims at configuring OpenFlow rules in the network using an SDN controller to perform SFC. Moreover, the focus will be on how to represent an SDN element and a service function chain in the legacy network NMS. The thesis also contains a discussion on what information should be exchanged between the management software and the controller. The management software used is called BECS, a system developed by Packetfront Software. Integrating SDN in BECS is done by creating a proof of concept, containing a full environment from the low level network elements to the NMS. By using a bottom-up approach for creating this proof of concept, the information that BECS is required to send to the SDN controller can be identified before designing and implementing the connection between these two entities. When sending the information, the NMS should be able to receive acknowledgement of successful information exchange or an error. However, when the proof of concept was created a problem arose on how to test and troubleshoot it. For this reason, a web Graphical User Interface (GUI) was created. This GUI shows the number of packets that have gone through a VNF. Because it is possible to see how many packets go through a VNF, one can see where a network issue occurs. The subsequent analysis investigates the impact of making such a GUI available for a network administrator and finds that the part of the network where the configuration error occurs can be narrowed down significantly. / Software Defined Networking (SDN) har blivit mer och mer populärt i kombination med Network Function Virtualization (NFV). SDN är en sätt för att göra ett nätverk mer programmerbart och dynamiskt. För att skapa ett homogent nätverk med detta koncept, behöver man dock ersätta traditionell utrustning med SDN utrustning som är dyr. För att stänga gapet mellan traditionella nätverk och SDN-världen, introducerar vi ett koncept med ett traditionell Network Management System (NMS) som är anslutet till en SDN-styrenhet för att utföra tjänsteprovisionering. På detta sätt kan NMS:et konfigurera både traditionella och SDN-nätverk, samt provisionera tjänster för kunderna medan nya SDN-funktioner möjliggörs i SDN-delen av nätverket. Den huvudsakliga tjänsten som vi vill lansera genom SDN är Service Function Chaining (SFC). Tjänsteprovisionering består av att konstruera en väg genom beställda tjänster, i detta fall Virtual Network Functions (VNFs). Detta examensarbete fokuserar huvusakligen på SDN-styrenheten och dess interaktion med NMS:et. Projektet syftar till att konfigurera OpenFlow regler i SDN-styrenheten för att utföra SFC. Dessutom fokuserar arbetet på hur man kan representera SDN-element och SFCs i ett traditionellt NMS. Vidare diskuteras vilken information som ska utbytas mellan NMS:et och SDNstyrenheten. NMS:et som ska vara användas är BECS, ett system utvecklat av Packetfront Software. Uppgiften löses genom att skapa ett proof of concept, som innehåller ett komplett system med alla komponenter från nätverkselement till NMS:et. Genom att använda en bottom-up-strategi för detta proof of concept kan informationen som BECS måste skicka till SDN styrenheten indentifieras, innan design och implementation av förbindelsen mellan enheterna kan utföras. När informationen är skickad ska NMS:et kunna hämta information om huruvida styrenheten fick informationen utan fel. Dock uppstår ett problem gällande hur man testar och felsöker detta proof of concept. Av denna anledning skapades ett web Graphical User Interface (GUI). Användargränssnittet visar antalet paket som går genom varje VNF, samt var i nätverket fel uppstår. Analysen undersöker hur stor effekten är för en nätverkadministrator och visar att området där fel kan uppstå begränsas avsevärt.

With the advent of Network Function Virtualization (NFV), network services that traditionally run on proprietary dedicated hardware can now be realized using Virtual Network Functions (VNFs) that are hosted on general-purpose commodity hardware. This new network paradigm offers a great flexibility to Internet service providers (ISPs) for efficiently operating their networks (collecting network statistics, enforcing management policies, etc.). However, introducing NFV requires an investment to deploy VNFs at certain network nodes (called VNF-nodes), which has to account for practical constraints such as the deployment budget and the VNF-node limited resources. While gradually transitioning to NFV, ISPs face the problem of where to efficiently introduce NFV; here, we measure the efficiency by the amount of traffic that can be served in an NFV-enabled network. This problem is non-trivial as it is composed of two challenging subproblems: 1) placement of VNF-nodes; 2) allocation of the VNF-nodes' resources to network flows. These two subproblems must be jointly considered to satisfy the objective of serving the maximum amount of traffic. We first consider this problem for the one-dimensional setting, where all network flows require one network function, which requires a unit of resource to process a unit of flow. In contrast to most prior work that often neglects either the budget constraint or the resource allocation constraint, we explicitly consider both of them and prove that accounting for them introduces several new challenges. Specifically, we prove that the studied problem is not only NP-hard but also non-submodular. To address these challenges, we introduce a novel relaxation method such that the objective function of the relaxed placement subproblem becomes submodular. Leveraging this useful submodular property, we propose two algorithms that achieve an approximation ratio of $\frac{1}{2}(1-1/e)$ and $\frac{1}{3}(1-1/e)$ for the original non-relaxed problem, respectively. Next, we consider the multi-dimensional setting, where flows can require multiple network functions, which can also require a different amount of each resource to process a unit of flow. To address the new challenges arising from the multi-dimensional setting, we propose a novel two-level relaxation method that allows us to draw a connection to the sequence submodular theory and utilize the property of sequence submodularity along with the primal-dual technique to design two approximation algorithms. Finally, we perform extensive trace-driven simulations to show the effectiveness of the proposed algorithms. While the NFV paradigm offers great flexibility to network operators for efficient management of their networks, VNF instances are typically more prone to error and more vulnerable to security threats compared with dedicated hardware devices. Therefore, the NFV paradigm also poses new challenges concerning failure resilience. That has motivated us to consider robustness with respect to the class of sequence submodular function maximization problem, which has a wide range of applications, including those in the NFV domain. Submodularity is an important property of set functions and has been extensively studied in the literature. It models set functions that exhibit a diminishing returns property, where the marginal value of adding an element to a set decreases as the set expands. This notion has been generalized to considering sequence functions, where the order of adding elements plays a crucial role and determines the function value; the generalized notion is called sequence (or string) submodularity. In this part of the dissertation, we study a new problem of robust sequence submodular maximization with cardinality constraints. The robustness is against the removal of a subset of elements in the selected sequence (e.g., due to malfunctions or adversarial attacks). Compared to robust submodular maximization for set function, new challenges arise when sequence functions are concerned. Specifically, there are multiple definitions of submodularity for sequence functions, which exhibit subtle yet critical differences. Another challenge comes from two directions of monotonicity: forward monotonicity and backward monotonicity, both of which are important to proving performance guarantees. To address these unique challenges, we design two robust greedy algorithms: while one algorithm achieves a constant approximation ratio but is robust only against the removal of a subset of contiguous elements, the other is robust against the removal of an arbitrary subset of the selected elements but requires a stronger assumption and achieves an approximation ratio that depends on the number of the removed elements. Finally, we consider important problems that arise in the production networks, where packets need to pass through an ordered set of network functions called Service Function Chains (SFC) before reaching the destination. We study the following problems: (1) How to find an SFC-constrained shortest path between any pair of nodes? (2) What is the achievable SFC-constrained maximum flow? We propose a transformation of the network graph to minimize the computational complexity of subsequent applications of any shortest path algorithm. Moreover, we formulate the SFC-constrained maximum flow problem as a fractional multicommodity flow problem and develop a combinatorial algorithm for a special case of practical interest. / Computer and Information Science

A datacenter hosts hundreds of thousands of servers and a huge amount of bandwidth is required to accommodate communication between thousands of servers. Several packet switched based datacenter architectures are proposed to cater the high bandwidth requirement using multilayer network topologies, however at the cost of increased network complexity and high power consumption. In recent years, the focus has shifted from packet switching to optical circuit switching to build the data center networks as it can support on demand connectivity and high bit rates with low power consumption. On the other hand, with the advent of Software Defined Networking (SDN) and Network Function Virtualization (NFV), the role of datacenters has become more crucial. It has increased the need of dynamicity and flexibility within a datacenter adding more complexity to datacenter networking. With NFV, service chaining can be achieved in a datacenter where virtualized network functions (VNFs) running on commodity servers in a datacenter are instantiated/terminated dynamically. A datacenter also needs to cater large capacity requirement as service chaining involves steering of large aggregated flows. Use of optical circuit switching in data center networks is quite promising to meet such dynamic and high capacity traffic requirements. In this thesis work, a novel and modular optical data center network (DCN) architecture that uses multi-directional wavelength switches (MD-WSS) is introduced. VNF service chaining use case is considered for evaluation of this DCN and the end-to-end service chaining problem is formulated as three inter-connected sub-problems: multiplexing of VNF service chains, VNFs placement in the datacenter and routing and wavelength assignment. This thesis presents integer linear programming (ILP) formulation and heuristics for solving these problems, and numerically evaluate them. / Ett datacenter inrymmer hundratusentals servrar och en stor mängd bandbredd krävs för att skicka data mellan tusentals servrar. Flera datacenter baserade på paketförmedlande arkitekturer föreslås för att tillgodose kravet på hög bandbredd med hjälp av flerskiktsnätverkstopologier, men på bekostnad av ökad komplexitet i nätverken och hög energiförbrukning. Under de senaste åren har fokus skiftat från paketförmedling till optisk kretsomkoppling for att bygga datacenternätverk som kan stödja på-begäran-anslutningar och höga bithastigheter med låg strömförbrukning. Å andra sidan, med tillkomsten av Software Defined Networking (SDN) och nätverksfunktionen Virtualisering (NFV), har betydelsen av datacenter blivit mer avgörande. Det har ökat behovet av dynamik och flexibilitet inom ett datacenter, vilket leder till storre komplexitet i datacenternätverken. Med NFV kan tjänstekedjor åstadkommas i ett datacenter, där virtualiserade nätverksfunktioner (VNFs) som körs på servrar i ett datacenter kan instansieras och avslutas dynamiskt. Ett datacenter måste också tillgodose kravet på stor kapacitet eftersom tjänstekedjan innebär styrning av stora aggregerade flöden. Användningen av optisk kretsomkoppling i datacenternätverk ser ganska lovande ut for att uppfylla sådana trafikkrav dynamik och hög kapacitet. I detta examensarbete, har en ny och modulär optisk datacenternätverksarkitektur (DCN) som använder flerriktningvåglängdsswitchar (MD-WSS) införs. Ett användningsfall av VNF-tjänstekedjor noga övervägd för utvärdering av denna DCN och end-to-end-servicekedjans problem formuleras som tre sammankopplade delproblem: multiplexering av VNF-servicekedjor, VNF placering i datacentret och routing och våglängd uppdrag. Denna avhandling presenterar heltalsprogrammering (ILP) formulering och heuristik för att lösa dessa problem och numeriskt utvärdera dem.

Avec la croissance exponentielle des demandes de service, les opérateurs ont déployé de nombreux équipements, et par conséquent, la gestion du réseau est devenue de plus en plus difficile et coûteuse. La virtualisation des fonctions réseau (NFV) a été proposée comme un nouveau paradigme pour réduire les coûts liés à l’acquisition et à la maintenance pour les réseaux de télécommunications. Dans ce travail de thèse, nous nous intéressons aux problèmes du chaînage des fonctions virtuelles (VNFs) qui combinent des décisions de localisation des VNFs et de routage des demandes. D'un point de vue d'optimisation, ce problème est une combinaison des problèmes de localisation (pour la partie d'installation des VNFs) et de conception de réseaux (pour la partie de routage). Ces deux problèmes ont été largement étudié dans la littérature. Cependant, leur combinaison représente des divers challenges en termes de modélisation et de résolution. Dans la première partie de cette thèse, nous considérons une version réaliste du problème du chaînage des VNFs (VNF-PR) afin de comprendre l'impact des différents aspects sur les coûts et les performances de gestion du réseau. Dans ce but, nous étendons le travail dans~\cite{Addis2015} en considérant des caractéristiques et des contraintes plus réalistes des infrastructures NFV et nous proposons un modèle de programmation linéaire et une heuristique mathématique pour le résoudre. Dans le but de mieux comprendre la structure du problème et ses propriétés, la deuxième partie de la thèse est orientée vers l'étude théorique du problème, où nous avons étudié une version compacte du problème du chaînage des VNFs. Nous fournissons des résultats sur la complexité de calcul sous divers cas de topologie et de capacité. Ensuite, nous proposons deux modèles et nous les testons sur un testbed avec plus de 100 instances différentes avec différents cas de capacité. Au final, nous abordons la scalabilité du problème en proposant des méthodes constructives et des méthodes heuristiques basées sur la programmation linéaire entière pour traiter efficacement des instances de taille grande (jusqu'à 60 nœuds et 1800 demandes). Nous montrons que les heuristiques proposées sont capables de résoudre efficacement des instances de taille moyenne (avec jusqu'à 30 nœuds et 1 000 demandes) de cas de capacité difficiles et de trouver de bonnes solutions pour les instances dures, où le modèle ne peut fournir aucune solution avec un temps de calcul limité. / Due to the exponential growth of service demands, telecommunication networks are populated with a large and increasing variety of proprietary hardware appliances, and this leads to an increase in the cost and the complexity of the network management. To overcome this issue, the NFV paradigm is proposed, which allows dynamically allocating the Virtual Network Functions (VNFs) and therefore obtaining flexible network services provision, thus reducing the capital and operating costs. In this thesis, we focus on the VNF Placement and Routing (VNF-PR) problem, which aims to find the location of the VNFs to allocate optimally resources to serve the demands. From an optimization point of view, the problem can be modeled as the combination of a facility location problem (for the VNF location and server dimensioning) and a network design problem (for the demands routing). Both problems are widely studied in the literature, but their combination represents, to the best of our knowledge, a new challenge. We start working on a realistic VNF-PR problem to understand the impact of different policies on the overall network management cost and performance. To this end, we extend the work in [1] by considering more realistic features and constraints of NFV infrastructures and we propose a linear programming model and a math-heuristic to solve it. In order to better understand the problem structure and its properties, in the second part of our work, we focus on the theoretical study of the problem by extracting a simplified, yet significant variant. We provide results on the computational complexity under different graph topology and capacity cases. Then, we propose two mathematical programming formulations and we test them on a common testbed with more than 100 different test instances under different capacity settings. Finally, we address the scalability issue by proposing ILP-based constructive methods and heuristics to efficiently deal with large size instances (with up to 60 nodes and 1800 demands). We show that our proposed heuristics can efficiently solve medium size instances (with up to 30 nodes and 1000 demands) of challenging capacity cases and provide feasible solutions for large size instances of the most difficult capacity cases, for which the models cannot find any solution even with a significant computational time.

O paradigma Network Function Virtualization (NFV) promete tornar as redes de computadores mais escaláveis e flexíveis, através do desacoplamento das funções de rede de hardware dedicado e fornecedor específico. No entanto, funções de rede computacionalmente intensivas podem ser difíceis de virtualizar sem degradação de desempenho. Neste contexto, Field-Programmable Gate Arrays (FPGAs) têm se mostrado uma boa opção para aceleração por hardware de funções de rede virtuais que requerem alta vazão, sem se desviar do conceito de uma infraestrutura NFV que visa alta flexibilidade. A avaliação de expressões regulares é um mecanismo importante e computacionalmente intensivo, usado para realizar Deep Packet Inpection, que pode ser acelerado por FPGA para atender aos requisitos de desempenho. Esta solução, no entanto, apresenta novos desafios em relação aos requisitos de confiabilidade. Particularmente para FPGAs baseados em SRAM, soft errors na memória de configuração são uma ameaça de confiabilidade significativa. Neste trabalho, apresentamos um mecanismo de tolerância a falhas abrangente para lidar com falhas de configuração na funcionalidade de módulos de avaliação de expressões regulares baseados em FPGA. Além disso, é introduzido um mecanismo de correção de erros que considera o posicionamento desses módulos no FPGA para reduzir o tempo de reparo do sistema, melhorando a confiabilidade e a disponibilidade. Os resultados experimentais mostram que a taxa de falha geral e o tempo de reparo do sistema podem ser reduzidos em 95% e 90%, respectivamente, com custos de área e performance admissíveis. / The Network Function Virtualization (NFV) paradigm promises to make computer networks more scalable and flexible by decoupling the network functions (NFs) from dedicated and vendor-specific hardware. However, network and compute intensive NFs may be difficult to virtualize without performance degradation. In this context, Field-Programmable Gate Arrays (FPGAs) have been shown to be a good option for hardware acceleration of virtual NFs that require high throughput, without deviating from the concept of an NFV infrastructure which aims at high flexibility. Regular expression matching is an important and compute intensive mechanism used to perform Deep Packet Inspection, which can be FPGA-accelerated to meet performance constraints. This solution, however, introduces new challenges regarding dependability requirements. Particularly for SRAM-based FPGAs, soft errors on the configuration memory are a significant dependability threat. In this work we present a comprehensive fault tolerance mechanism to deal with configuration faults on the functionality of FPGA-based regular expression matching engines. Moreover, a placement-aware scrubbing mechanism is introduced to reduce the system repair time, improving the system reliability and availability. Experimental results show that the overall failure rate and the system mean time to repair can be reduced in 95% and 90%, respectively, with manageable area and performance costs.

O paradigma Network Function Virtualization (NFV) promete tornar as redes de computadores mais escaláveis e flexíveis, através do desacoplamento das funções de rede de hardware dedicado e fornecedor específico. No entanto, funções de rede computacionalmente intensivas podem ser difíceis de virtualizar sem degradação de desempenho. Neste contexto, Field-Programmable Gate Arrays (FPGAs) têm se mostrado uma boa opção para aceleração por hardware de funções de rede virtuais que requerem alta vazão, sem se desviar do conceito de uma infraestrutura NFV que visa alta flexibilidade. A avaliação de expressões regulares é um mecanismo importante e computacionalmente intensivo, usado para realizar Deep Packet Inpection, que pode ser acelerado por FPGA para atender aos requisitos de desempenho. Esta solução, no entanto, apresenta novos desafios em relação aos requisitos de confiabilidade. Particularmente para FPGAs baseados em SRAM, soft errors na memória de configuração são uma ameaça de confiabilidade significativa. Neste trabalho, apresentamos um mecanismo de tolerância a falhas abrangente para lidar com falhas de configuração na funcionalidade de módulos de avaliação de expressões regulares baseados em FPGA. Além disso, é introduzido um mecanismo de correção de erros que considera o posicionamento desses módulos no FPGA para reduzir o tempo de reparo do sistema, melhorando a confiabilidade e a disponibilidade. Os resultados experimentais mostram que a taxa de falha geral e o tempo de reparo do sistema podem ser reduzidos em 95% e 90%, respectivamente, com custos de área e performance admissíveis. / The Network Function Virtualization (NFV) paradigm promises to make computer networks more scalable and flexible by decoupling the network functions (NFs) from dedicated and vendor-specific hardware. However, network and compute intensive NFs may be difficult to virtualize without performance degradation. In this context, Field-Programmable Gate Arrays (FPGAs) have been shown to be a good option for hardware acceleration of virtual NFs that require high throughput, without deviating from the concept of an NFV infrastructure which aims at high flexibility. Regular expression matching is an important and compute intensive mechanism used to perform Deep Packet Inspection, which can be FPGA-accelerated to meet performance constraints. This solution, however, introduces new challenges regarding dependability requirements. Particularly for SRAM-based FPGAs, soft errors on the configuration memory are a significant dependability threat. In this work we present a comprehensive fault tolerance mechanism to deal with configuration faults on the functionality of FPGA-based regular expression matching engines. Moreover, a placement-aware scrubbing mechanism is introduced to reduce the system repair time, improving the system reliability and availability. Experimental results show that the overall failure rate and the system mean time to repair can be reduced in 95% and 90%, respectively, with manageable area and performance costs.