With the advent of Smart Cities, Intelligent Transport System (ITS) has become an efficient way of offering an accessible, safe, and sustainable transportation system. Utilizing advances in Information and Communication Technology (ICT), ITS can maximize the capacity of existing transportation system without building new infrastructure. However, in spite of these technical feasibilities and significant performance-cost ratios, the deployment of ITS is limited in the real world because of several challenges associated with its architectural design. This thesis studies how to design a highly flexible and deployable architecture for ITS, which can utilize the recent technologies such as - cloud computing and the publish/subscribe communication model. In particular, our aim is to offer an ITS infrastructure which provides the opportunity for transport authorities to allocate on-demand computing resources through virtualization technology, and supports a wide range of ITS applications. We propose to use an Infrastructure as a Service (IaaS) model to host large-scale ITS applications for transport authorities in the cloud, which reduces infrastructure cost, improves management flexibility and also ensures better resource utilization. Moreover, we use a publish/subscribe system as a building block for developing a low latency ITS application, which is a promising technology for designing scalable and distributed applications within the ITS domain. Although cloud-based architectures provide the flexibility of adding, removing or moving ITS services within the underlying physical infrastructure, it may be difficult to provide the required quality of service (QoS) which decrease application productivity and customer satisfaction, leading to revenue losses. Therefore, we investigate the impact of service mobility on related QoS in the cloud-based infrastructure. We investigate different strategies to improve performance of a low latency ITS application during service mobility such as utilizing multiple paths to spread network traffic, or deploying recent queue management schemes. Evaluation results from a private cloud testbed using OpenStack show that our proposed architecture is suitable for hosting ITS applications which have stringent performance requirements in terms of scalability, QoS and latency. / Baksidestext: Intelligent Transport System (ITS) can utilize advances in Information and Communication Technology (ICT) and maximize the capacity of existing transportation systems without building new infrastructure. However, in spite of these technical feasibilities and significant performance-cost ratios, the deployment of ITS is limited in the real world because of several challenges associated with its architectural design.  This thesis studies how to design an efficient deployable architecture for ITS, which can utilize the advantages of cloud computing and the publish/subscribe communication model. In particular, our aim is to offer an ITS infrastructure which provides the opportunity for transport authorities to allocate on-demand computing resources through virtualization technology, and supports a wide range of ITS applications. We propose to use an Infrastructure as a Service (IaaS) model to host large-scale ITS applications, and to use a publish/subscribe system as a building block for developing a low latency ITS application. We investigate different strategies to improve performance of an ITS application during service mobility such as utilizing multiple paths to spread network traffic, or deploying recent queue management schemes. / <p>Artikel 4 Network Centric Performance Improvement for Live VM Migration finns i avhandlingen som manuskript. Nu publicerat konferenspaper. </p>

Afin de réduire l’intervalle de temps nécessaire entre la publication de l’information sur le Web et sa consultation par les utilisateurs, les sites Web reposent sur le principe de la Syndication Web. Les fournisseurs d’information diffusent les nouvelles informations à travers des flux RSS auxquels les utilisateurs intéressés peuvent s’abonner. L’objectif de la thèse est de proposer un système de notification passant à l’échelle du Web, prenant en considération le grand nombre d’utilisateurs et le débit élevé d’items. Nous proposons un index basé sur les mots-clés des requêtes utilisateurs permettant de retrouver ceux-ci dans les items des flux. Trois structures d’indexation de souscriptions sont présentées. Un modèle analytique pour estimer le temps de traitement et l’espace mémoire de chaque structure est détaillé. Nous menons une étude expérimentale approfondie de l’impact de plusieurs paramètres sur ces structures. Pour les souscriptions jamais notifiées, nous adaptons les index étudiés pour prendre en considération leur satisfaction partielle. Afin de réduire le nombre d’items reçus par l’utilisateur, nous intégrons une deuxième phase de filtrage par nouveauté et diversité considérant l’ensemble d’items déjà reçus par l’utilisateur. / Based on a Publish/Subscribe paradigm, Web Syndication formats such as RSS have emerged as a popular means for timely delivery of frequently updated Web content. According to these formats, information publishers provide brief summaries of the content they deliver on the Web, while information consumers subscribe to a number of RSS feeds and get informed about newly published items. The goal of this thesis is to propose a notification system which scales on the Web. To deal with this issue, we should take into account the large number of users on the Web and the high publication rate of items. We propose a keyword-based index for user subscriptions to match it on the fly with incoming items. We study three indexing techniques for user subscriptions. We present analytical models to estimate memory requirements and matching time. We also conduct a thorough experimental evaluation to exhibit the impact of critical workload parameters on these structures. For subscriptions which are never notified, we adapt the indexes to support a partial matching between subscriptions and items. We integrate a diversity and novelty filtering step in our system in order to decrease the number of notified items for short subscriptions. This filtering is based on the set of items already received by the user.

With the rapid development of the Internet of Things, Wireless Sensor Networks (WSNs) are deployed increasingly all over the world, providing data that can help increase sustainable development. Currently in Uppsala, Sweden, the GreenIoT project monitors air pollution by using WSNs. The resource constrained nature of WSNs demand that special care is taken in the design of communication models and communication protocols. The publish-subscribe (pub/sub) model suits WSNs very well since it puts an intermediary the broker server between sensor nodes and clients, thus alleviating the workload of the sensor nodes. The IETF (Internet Engineering Task Force) is currently in the process of standardizing a pub/sub extension to the Constrained Application Protocol (CoAP). Since the extension is such a recent addition to CoAP and not yet standardized, there are very few actual implementations of it and little is known of how it would work in practice. The GreenIoT project is considering replacement of their current pub/sub broker with the CoAP pub/sub broker since its underlying implementation is likely to be more energy efficient and the standardizing organization behind CoAP is the well-esteemed IETF. On a general level, this report offers an investigation of the problems and challenges faced when implementing the CoAP pub/sub extension with respect to design choices, implementation and protocol ambiguities. More specifically, a CoAP pub/sub broker is implemented for the GreenIoT project. By means of carefully analyzing the CoAP protocol and CoAP pub/sub draft as well as other necessary protocols, then proceeding to make decisions of what programming language to use as well as what existing CoAP library to use, a broker server was implemented and tested iteratively as the work proceeded. The implementation gave rise to several questions regarding the pub/sub draft which are also discussed in the report. / Den hastiga utvecklingen av Sakernas Internet över hela världen har medfört ett ökat användande av trådlösa sensornätverk vars datainsamling kan bidra till en mer hållbar utveckling. För närvarande använder sig GreenIoT-projektet i Uppsala av trådlösa sensornätverk för att övervaka halterna av luftföroreningar. Resursbegränsningarna för dylika nätverk medför att särskild hänsyn måste tas vid design av såväl kommunikationsmodeller som kommunikationsprotokoll. Modellen Publicera-Prenumerera (pub/pre) passar ypperligt för trådlösa sensornätverk då en mellanhand placeras mellan klient och server en s.k. broker vilket får den positiva effekten att att sensornoderna avlastas. För närvarande är IETF (Internet Engineering Task Force) i färd med att standardisera en pub/pre-utvidgning av det redan standardiserade CoAP (Constrained Application Protocol). Eftersom att utvidgningen är så pass ny finns ytterst få implementationer av den och man vet därmed väldigt lite om hur den faktiskt fungerar i praktiken. GreenIoT-projektet överväger att ersätta sin nuvarande pub/pre-broker med en CoAP pub/pre-broker eftersom att energianvändningen kan antas bli lägre samt att standardiseringsorganisationen bakom CoAP är det välrenommerade IETF. Sett ur ett större perspektiv erbjuder denna rapport en undersökning av de problem och utmaningar man ställs inför vid implementation av CoAP pub/pre-utvidgningen med avseende på designval, implementationsval, och protokolltvetydigheter. Mer konkret implementeras en CoAP pub/pre-broker åt GreenIoT-projektet. Genom att först noggrant analysera CoAP-protokollet, CoAP pub/pre-utkastet, liksom andra nödvändiga protokoll, för att därefter bestämma vilket programmeringsspråk och vilket existerande CoAP-bibliotek som skulle användas, implementerades en broker server som testades iterativt under processens gång. Ett flertal frågor som uppstod rörande pub/pre-utkastet presenteras och diskuteras i rapporten.

Le parking intelligent, permettant aux conducteurs d'accéder aux informations de stationnement sur leurs appareils mobiles, réduit les difficultés des usagers. Tout d'abord, nous mettons en lumière la manière de recueillir les informations de parking en introduisant une architecture de réseaux de capteurs multi-saut, et les modèles d'intensité applicative en examinant la probabilité d'arrivées et de départs de véhicules. Puis nous étudions la stratégie de déploiement des réseaux de capteurs et définissons un problème multi-objectifs, puis nous le résolvons sur deux cartes de parking réelles. Ensuite, nous définissons un service Publish-Subscribe pour fournir aux conducteurs des informations pertinentes. Nous illustrons le système dans des réseaux véhiculaires et mobiles et soulignons l'importance du contenu et du contexte du message au conducteur. Afin d'évaluer la résilience du système, nous proposons un modèle Publish-Subscribe étendu et nous l'évaluons dans différentes circonstances imprévues. Notre travail est basé sur la prémisse que les capteurs de parking sont déployés à une grande échelle dans la ville. Nous considérons une vue d'ensemble des services urbains du point de vue de la municipalité. Ainsi, nous faisons la lumière sur deux thèmes principaux: la collecte d'informations sur le déploiement de capteurs et un modèle étendu de Publish-Subscribe. Notre travail donne un guide avant de démarrer un projet de parking intelligent ou tout service urbain similaire en temps réel. Il fournit également une plate-forme d'évaluation valable pour tester des jeux de données plus réalistes, comme des traces de véhicules ou de trafic réseau. / Smart parking, allowing drivers to access parking information through their smart-phone, is proposed to ease drivers' pain. We first spotlight the manner to collect parking information by introducing the multi-hop sensor network architecture, and how the network is formed. We then introduce the traffic intensity models by looking at the vehicle's arrival and departure probabilities, following the heavy-tailed distribution. We study the deployment strategy of wireless on-street parking sensor layouts. We define a multiple-objective problem and solve it with two real street parking maps. In turn, we present a Publish-Subscribe service system to provide good parking information to drivers. We illustrate the system with a vehicular network and point out the importance of content and context of a driver’s message. To evaluate the resilience, we propose an extended Publish-Subscribe model, and evaluate it under different unforeseen circumstances. Our work is based on the premise that large-scale parking sensors are deployed in the city. We look at the whole picture of urban service from viewpoint of the municipality. As such, we shed light on two main topics: the information collection on sensor deployment and an extended version of Publish-Subscribe messaging paradigm. Our work gives a guideline from network-related perspectives for city before launching a smart parking or any similar real-time urban service. It also provides a meaningful evaluation platform for testing more realistic datasets, such as real vehicle traces or network traffic.

Distributed event processing systems suffer from poor scalability and inefficient resource usage caused by load distributions typical in real-world applications. The results of these shortcomings are availability issues, poor system performance, and high operating costs. This thesis proposes three remedies to solve these limitations in content-based publish/subscribe, which is a practical realization of an event processing system. First, we present a load balancing algorithm that relocates subscribers to distribute load and avoid overloads. Second, we propose publisher relocation algorithms that reduces both the load imposed onto brokers and delivery delay experienced by subscribers. Third, we present ``green" resource allocation algorithms that allocate as few brokers as possible while maximizing their resource usage efficiency by reconfiguring the publishers, subscribers, and the broker topology. We implemented and evaluated all of our approaches on an open source content-based publish/subscribe system called PADRES and evaluated them on SciNet, PlanetLab, a cluster testbed, and in simulations to prove the effectiveness of our solutions. Our evaluation findings are summarized as follows. One, the proposed load balancing algorithm is effective in distributing and balancing load originating from a single server to all available servers in the network. Two, our publisher relocation algorithm reduces the average input load of the system by up to 68%, average broker message rate by up to 85%, and average delivery delay by up to 68%. Three, our resource allocation algorithm reduces the average broker message rate even further by up to 92% and the number of allocated brokers by up to 91%.

Distributed event processing systems suffer from poor scalability and inefficient resource usage caused by load distributions typical in real-world applications. The results of these shortcomings are availability issues, poor system performance, and high operating costs. This thesis proposes three remedies to solve these limitations in content-based publish/subscribe, which is a practical realization of an event processing system. First, we present a load balancing algorithm that relocates subscribers to distribute load and avoid overloads. Second, we propose publisher relocation algorithms that reduces both the load imposed onto brokers and delivery delay experienced by subscribers. Third, we present ``green" resource allocation algorithms that allocate as few brokers as possible while maximizing their resource usage efficiency by reconfiguring the publishers, subscribers, and the broker topology. We implemented and evaluated all of our approaches on an open source content-based publish/subscribe system called PADRES and evaluated them on SciNet, PlanetLab, a cluster testbed, and in simulations to prove the effectiveness of our solutions. Our evaluation findings are summarized as follows. One, the proposed load balancing algorithm is effective in distributing and balancing load originating from a single server to all available servers in the network. Two, our publisher relocation algorithm reduces the average input load of the system by up to 68%, average broker message rate by up to 85%, and average delivery delay by up to 68%. Three, our resource allocation algorithm reduces the average broker message rate even further by up to 92% and the number of allocated brokers by up to 91%.

Just as packet switched networks constituted a major breakthrough in our perception of the information exchange in computer networks so have the decoupling properties of publish/subscribe systems revolutionized the way we look at networking in the context of large scale distributed systems. The decoupling of the components of publish/subscribe systems in time, space and synchronization has created an appealing platform for the asynchronous information exchange among anonymous information producers and consumers. Moreover, the content-based nature of publish/subscribe systems provides a great degree of flexibility and expressiveness as far as construction of data flows is considered. However, a number of challenges and not yet addressed issued still exists in the area of the publish/subscribe systems. One active area of research is directed toward the problem of the efficient content delivery in the content-based publish/subscribe networks. Routing of the information based on the information itself, instead of the explicit source and destination addresses poses challenges as far as efficiency and processing times are concerned. Simultaneously, due to their decoupled nature, publish/subscribe systems introduce new challenges with respect to issues related to dependability and fail-awareness. This thesis seeks to advance the field of research in both directions. First it shows the design and implementation of routing algorithms based on the end-to-end systems design principle. Proposed routing algorithms obsolete the need to perform content-based routing within the publish/subscribe network, pushing this task to the edge of the system. Moreover, this thesis presents a fail-aware approach towards construction of the content-based publish/subscribe system along with its application to the creation of the soft state publish/subscribe system. A soft state publish/subscribe system exposes the self stabilizing behavior as far as transient timing, link and node failures are concerned. The result of this thesis is a family of the XSiena content-based publish/subscribe systems, implementing the proposed concepts and algorithms. The family of the XSiena content-based publish/subscribe systems has been a subject to rigorous evaluation, which confirms the claims made in this thesis.

Modern hardware has the potential to play a central role in scalable data management systems. A realization of this potential arises in the context of indexing queries, a recurring theme in real-time data analytics, targeted advertising, algorithmic trading, and data-centric workflows, and of indexing data, a challenge in multi-version analytical query processing. To enhance query and data indexing, in this thesis, we present an efficient, extensible, and hardware-aware indexing kernel. This indexing kernel rests upon novel data structures and (parallel) algorithms that utilize the capabilities offered by modern hardware, especially abundance of main memory, multi-core architectures, hardware accelerators, and solid state drives. This thesis focuses on presenting our query indexing techniques to cope with processing queries in data-intensive applications that are susceptible to ever increasing data volume and velocity. At the core of our query indexing kernel lies the BE-Tree family of memory-resident indexing structures that scales by overcoming the curse of dimensionality through a novel two-phase space-cutting technique, an effective Top-k processing, and adaptive parallel algorithms to operate directly on compressed data (that exploits the multi-core architecture). Furthermore, we achieve line-rate processing by harnessing the unprecedented degrees of parallelism and pipelining only available through low-level logic design using FPGAs. Finally, we present a comprehensive evaluation that establishes the superiority of BE-Tree in comparison with state-of-the-art algorithms. In this thesis, we further expand the scope of our indexing kernel and describe how to accelerate analytical queries on (multi-version) databases by enabling indexes on the most recent data. Our goal is to reduce the overhead of index maintenance, so that indexes can be used effectively for analytical queries without being a heavy burden on transaction throughput. To achieve this end, we re-design the data structures in the storage hierarchy to employ an extra level of indirection over solid state drives. This indirection layer dramatically reduces the amount of magnetic disk I/Os that is needed for updating indexes and localizes the index maintenance. As a result, by rethinking how data is indexed, we eliminate the dilemma between update vs. query performance and reduce index maintenance and query processing cost substantially.

Modern hardware has the potential to play a central role in scalable data management systems. A realization of this potential arises in the context of indexing queries, a recurring theme in real-time data analytics, targeted advertising, algorithmic trading, and data-centric workflows, and of indexing data, a challenge in multi-version analytical query processing. To enhance query and data indexing, in this thesis, we present an efficient, extensible, and hardware-aware indexing kernel. This indexing kernel rests upon novel data structures and (parallel) algorithms that utilize the capabilities offered by modern hardware, especially abundance of main memory, multi-core architectures, hardware accelerators, and solid state drives. This thesis focuses on presenting our query indexing techniques to cope with processing queries in data-intensive applications that are susceptible to ever increasing data volume and velocity. At the core of our query indexing kernel lies the BE-Tree family of memory-resident indexing structures that scales by overcoming the curse of dimensionality through a novel two-phase space-cutting technique, an effective Top-k processing, and adaptive parallel algorithms to operate directly on compressed data (that exploits the multi-core architecture). Furthermore, we achieve line-rate processing by harnessing the unprecedented degrees of parallelism and pipelining only available through low-level logic design using FPGAs. Finally, we present a comprehensive evaluation that establishes the superiority of BE-Tree in comparison with state-of-the-art algorithms. In this thesis, we further expand the scope of our indexing kernel and describe how to accelerate analytical queries on (multi-version) databases by enabling indexes on the most recent data. Our goal is to reduce the overhead of index maintenance, so that indexes can be used effectively for analytical queries without being a heavy burden on transaction throughput. To achieve this end, we re-design the data structures in the storage hierarchy to employ an extra level of indirection over solid state drives. This indirection layer dramatically reduces the amount of magnetic disk I/Os that is needed for updating indexes and localizes the index maintenance. As a result, by rethinking how data is indexed, we eliminate the dilemma between update vs. query performance and reduce index maintenance and query processing cost substantially.

[pt] Esta dissertação apresenta uma abordagem para a coordenação de enxames de Veículos Aéreos Não Tripulados (VANT), baseada na comunicação via Internet provida pelas redes de telefonia móvel. Um grande número de atividades pode ser coberto com uma missão conjunta de VANTs voando em formação, como a vigilância e monitoramento de grandes eventos, tarefas de busca e salvamento, controle de pestes agrárias, fiscalização e conservação florestal, inspeção de oleodutos e redes de distribuição elétrica ou mesmo em missões militares de ataque e reconhecimento. A coordenação de enxame de VANTs pode ser fatorada na comunicação entre os membros do enxame e o algoritmo de controle e interrelacionamento entre os membros. A solução proposta consiste no uso de smartphones acoplados a VANTs capazes de prover o nível necessário de comunicação confiável sobre a Internet móvel, e processar o algoritmo proposto para a coordenação dos enxames de VANTs. Experimentos foram feitos através da emulação de VANTs e redes de dados de telefonia que demostraram a eficácia do algoritmo proposto e analisam o impacto da latência de rede e da velocidade dos VANTs sobre a precisão da coordenação de movimento de enxames. / [en] This work presents an approach to coordinate swarms of Unmanned Aerial Vehicles (UAV) based on Internet communication provided by mobile phone networks. Several activities can be done by several UAVs flying in formation, such as surveillance and monitoring of mass events, search and rescue tasks, control of agricultural pests, monitoring and forest conservation, inspection of pipelines and electricity distribution networks or even military attack and recognition missions. Coordination of UAVs swarm can be branch in two sub-problems: communication between members of the swarm and the algorithm that controls members behaviors regarding their movements. The proposed solution assumes the use of a smartphone coupled with each UAV of the swarm, in order to provide the required level of reliable communication on the mobile Internet and run the proposed algorithm for the coordination of swarms of UAVs. Experiments were performed with emulated UAVs and WAN mobile networks. The results have demonstrated the effectiveness of the proposed algorithm, and have shown the influence of the network latency and the UAV speeds on the accuracy of the movement coordination in the swarms.