The application offloading problem for Mobile Cloud Computing aims at improving the mobile user experience by leveraging the resources of the cloud. The execution of the mobile application is offloaded to the cloud, saving energy at the mobile device or speeding up the execution of the application. We improve the accuracy and performance of application offloading solutions in three main directions. First, we propose a novel fine-grained application model that supports complex module dependencies such as sequential, conditional and parallel module executions. The model also allows for multiple offloading decisions that are tailored towards the current application, network, or user contexts. As a result, the model is more precise in capturing the structure of the application and supports more complex offloading solutions. Second, we propose three cost models, namely, average-based, statistics-based and interval-based cost models, defined for the proposed application model. The average-based approach models each module cost by the expected cost value, and the expected cost of the entire application is estimated considering each of the three module dependencies. The novel statistics-based cost model employs Cumulative Distribution Function (CDFs) to represent the costs of the modules and of the mobile application, which is estimated considering the cost and dependencies of the modules. This cost model opens the doors for new statistics-based optimization functions and constraints whereas the state of the art only support optimizations based on the average running cost of the application. Furthermore, this cost model can be used to perform statistical analysis of the performance of the application in different scenarios such as varying network data rates. The last cost model, the interval-based, represents the module costs via intervals in order to addresses the cost uncertainty while having lower requirements and computational complexity than the statistics-based model. The cost of the application is estimated as an expected maximum cost via a linear optimization function. Finally, we present offloading decision algorithms for each cost model. For the average-based model, we present a fast optimal dynamic programming algorithm. For the statistics-based model, we present another fast optimal dynamic programming algorithm for the scenario where the optimization function meets specific properties. Finally, for the interval-based cost model, we present a robust formulation that solves a linear number of linear optimization problems. Our evaluations verify the accuracy of the models and show higher cost savings for our solutions when compared to the state of the art.

No description available.

Mobile cloud gaming (MCG) is an emerging concept which aims to deliver video games on-demand to users with the use of cloud technologies. Cloud technology allows the offloading of computation from a less powerful user device or thin client to more robust cloud servers to minimize power consumption and provide additional cloud services such as storage. MCG is therefore very helpful that can reduce the costs of expensive hardware, but the challenge is that it requires a high Quality of Service (QoS) in order to stream and play the games where the users have a high Quality of Experience (QoE). The goal of the study is to investigate how users' QoE is affected by network conditions while playing MCG and compare the results from a previous study. A testbed was made in order to conduct subjective tests where users are going to play Counter Strike: Global Offensive (CS: GO) on a smartphone using Steam Remote Play. The testbed consists of a router, tablet, smartphone, headset, Xbox controller, USB-C multi-port adapter and four different PC's. Participants on campus, both students and non-students, were invited to participate in the experiment. A total of 24 participants completed the tests; however, results from two participants were excluded due to software issues. There were 23 network conditions that was tested for each user and included factors such as round-trip time (RTT), packet losses, bursty jitter, random jitter or combinations of different factors. A multi-platform tool, ALTRUIST, was used to control the applications and facilitate the data collection from the devices and NetEm changed the network conditions. The results showed that the network condition [bj(rtt200i15)] had the highest mean opinion score (MOS) of the QoE of 4.5 for the users with 200 milliseconds of bursty jitter every 15 seconds. The worst network condition tested with the lowest QoE rating of 1.4 was network condition [rtt25pl12] that had 25 milliseconds of RTT and 12% packet losses. There were differences between the male and female participants where the MOS of the QoE results was significantly higher with up to 1.5 MOS QoE rating differences for the females compared to the males in network conditions with RTT with packet losses. However, the sample size was low with only 5 female participants compared to 18 male participants. The MOS of the QoE results separating play time under 10 hours per week and 10 or more hours per week showed no significant changes, where the largest QoE rating difference was 0.5 points. Network condition [rtt25pl12] and [rtt2pl35] had the largest differences in the MOS QoE ratings compared to the previous study, while both was not compared to the same corresponding network condition. The largest difference comparing the same network condition to the previous study was network condition [bj(rtt200i15)] with a difference of 1.1 points higher in the MOS QoE rating.

No description available.

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2017-04-25T12:03:54Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertacao_AdrianoHenrique.pdf: 1956295 bytes, checksum: 38ce5d73db0d44416c8653e58120f11c (MD5) / Made available in DSpace on 2017-04-25T12:03:55Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertacao_AdrianoHenrique.pdf: 1956295 bytes, checksum: 38ce5d73db0d44416c8653e58120f11c (MD5) Previous issue date: 2016-08-29 / Apesar da grande evolução tecnológica nos hardwares dos dispositivos móveis e nas redes sem fio, ainda existem grandes limitações nesses dispositivos em termos de processamento, capacidade de armazenamento e autonomia de energia, quando comparados aos desktops e servidores. O paradigma de Computação em Nuvem Móvel (MCC – Mobile Cloud Computing) permite estender os recursos computacionais dos dispositivos móveis através da utilização das técnicas de offloading computacional possibilitando um melhor desempenho as aplicações e uma redução no consumo das baterias dos dispositivos. Entretanto, a técnica de offloading nem sempre traz benefícios para o dispositivo móvel em situações de constante mobilidade do usuário, já que cada mudança de rede requer que o processo de offloading seja refeito. Esta dissertação propõe uma solução para otimizar o consumo de energia e o tempo de resposta durante as operações de offloading computacional quando o dispositivo muda de ponto de acesso. A proposta considera um esquema de gerenciamento de mobilidade baseado em Software Defined Networking (SDN) e técnica de caching remoto, que permite ao usuário receber o resultado do offloading no próximo acesso à rede, mesmo que esse fique desconectado por um longo período. A solução foi implementada em um testbed WiFi, com acesso ao ambiente MCC utilizando cloudlet baseada na plataforma OpenStack e integrada ao controlador SDN OpenDaylight. O consumo de energia obtido pela proposta que utiliza SDN/OpenFlow para o gerenciamento de mobilidade chegou a ser 11,33 vezes menor e a velocidade de processamento foi 3,23 vezes maior que do ambiente tradicional. O sistema de caching remoto, apesar de se mostrar útil em relação à rápida entrega dos resultados já processados, elevou consideravelmente o consumo de energia da bateria. A técnica de caching remoto é indicada para os casos nos quais a aplicação envia à cloudlet um grande volume de dados para ser processado e o nível da bateria do dispositivo encontra-se em estado não crítico ou quando o usuário enfrenta um longo período sem comunicação com a cloudlet. / Although the great technological evolution in the mobile devices hardware and wireless networks, remains significant limitations of these devices regarding processing, storage, and energy, when compared to desktops and servers. The paradigm of Mobile Cloud Computing (MCC) allows to extend the computational resources of the mobile devices through the use of computational offloading techniques, achieving a better performance on the part of the applications and a reduction in the battery consumption of the devices. The offloading technique does not always bring benefits to a mobile device in situations of high mobility since each network change requires the execution of the offloading process. This dissertation proposes a solution to optimize energy consumption and response times during the computational offloading operations when the device change of access points (AP). To this end, the proposal considers for such, a mobility management scheme based on SDN (Software Defined Networking) and a remote caching technique, that allows the user to receive the result from offloading in the next AP, even if he stays disconnected for an extended period. The solution was implemented in one Wi-Fi testbed, with access to the MCC environment using cloudlet based on the OpenStack platform and integrated with the OpenDaylight SDN controller. The achieved reduction of energy consumption for the mobility management proposal arrived to be 11.33 times lower, and the processing speed was 3.23 times bigger that of the traditional environment. The remote caching system, although useful in fast delivering the already processed results, considerably raised the battery energy consumption. Thus, the applicability of remote caching limits it to the cases where the application sends to the cloudlet an enormous volume of data to be processed and the battery level of the device is not critical or when the user faces an extended period without communication with the cloudlet.

De nos jours, nous assistons à une énorme avancée des applications collaboratives mobiles. Ces applications tirentparti de la disponibilité croissante des réseaux de communication et de l’évolution impressionnante des dispositifsmobiles. Cependant, même avec un développement en accélération, ils demeurent toujours pauvres en ressources(une courte durée de vie des batteries et une connexion réseau instable) et moins sécurisés. Dans le cadre de notretravail, nous proposons une nouvelle approche basée sur le déploiement des tâches de collaboration mobile versle cloud. La gestion d’une virtualisation efficace assurant la continuité de la collaboration pour des réseaux pairà-pair est une tâche très difficile. En effet, l’aspect dynamique des groupes (où les utilisateurs peuvent joindre,quitter ou changer de groupes) ainsi qu’une vulnérabilité aux pannes peuvent affecter la collaboration. En outre,la conception de telles applications doit prendre en compte l’hétérogénéité des environnements cloud et mobile.Contrairement aux travaux existants , nous proposons une architecture réutilisable de haut niveau basée sur les patronsde conception et qui peut être facilement adaptée à plusieurs environnements clouds et mobiles hétérogènes.Nos modèles ont été utilisés comme base pour la conception de : (i) MidBox, une plate-forme virtuelle pour exécuterdes applications collaboratives mobiles sur un cloud privé et (ii) MobiRDF, un service de cloud décentralisépour la manipulation en temps réel des connaissances via des documents RDF partagés. / Nowadays we assist to an enormous progress of mobile collaborative applications. These applications take advantage of the increasing availability of communication networks and the impressive evolution of mobile devices. However, even with a developing acceleration, they are still poor in resources (short life of batteries andunstable network connections) and less secure. In the context of our work, we propose a new approach based on the deployment of mobile collaboration tasks to the cloud. The management of efficient virtualization ensuring continuity of collaboration in peer-to-peer networks is a very difficult task. Indeed, the dynamic aspect of the groups (where users can join, leave or change groups) and a vulnerability to failures can affect the collaboration.In addition, the design of such applications must consider the heterogeneity of cloud and mobile environments.Unlike existing works, we propose a reusable high-level architecture based on patterns design, which can be easily adapted to heterogeneous clouds and mobile environments. Our models have been used as basis for the design of:(i) MidBox, a virtual platform for running mobile collaborative applications on a private cloud and (ii) MobiRDFa decentralized cloud service for real-time manipulation of knowledge via shared RDF documents.

Cette thèse apporte des contributions à la modélisation, compréhension ainsi qu’à la communication efficace d’information dans les réseaux dynamiques peuplant la périphérie de l’Internet. Par réseaux dynamiques, nous signifions les réseaux pouvant être modélisés par des graphes dynamiques dans lesquels noeuds et liens évoluent temporellement. Dans la première partie de la thèse, nous proposons un nouveau modèle de mobilité - STEPS - qui permet de capturer un large spectre de comportement de mobilité humains. STEPS mets en oeuvre deux principes fondamentaux de la mobilité humaine : l’attachement préférentiel à une zone de prédilection et l’attraction vers une zone de prédilection. Nous proposons une modélisation markovienne de ce modèle de mobilité. Nous montrons que ce simple modèle paramétrique est capable de capturer les caractéristiques statistiques saillantes de la mobilité humaine comme la distribution des temps d’inter-contacts et de contacts. Dans la deuxième partie, en utilisant STEPS, nous analysons les propriétés comportementales et structurelles fondamentales des réseaux opportunistes. Nous redéfinissons dans le contexte des réseaux dynamiques la notion de structure petit monde et montrons comment une telle structure peut émerger. En particulier, nous montrons que les noeuds fortement dynamiques peuvent jouer le rôle de ponts entre les composants déconnectés, aident à réduire significativement la longueur du chemin caractéristique du réseau et contribuent à l’émergence du phénomène petit-monde dans les réseaux dynamiques. Nous proposons une façon de modéliser ce phénomène sous STEPS. À partir d’un réseau dynamique régulier dans lequel les noeuds limitent leur mobilité à leurs zones préférentielles respectives. Nous recablons ce réseau en injectant progressivement des noeuds nomades se déplaçant entre plusieurs zones. Nous montrons que le pourcentage de tels nœuds nomades est de 10%, le réseau possède une structure petit monde avec un fort taux de clusterisation et un faible longueur du chemin caractéristique. La troisième contribution de cette thèse porte sur l’étude de l’impact du désordre et de l’irrégularité des contacts sur la capacité de communication d’un réseau dynamique. Nous analysons le degré de désordre de réseaux opportunistes réels et montrons que si exploité correctement, celui-ci peut améliorer significativement les performances du routage. Nous introduisons ensuite un modèle permettant de capturer le niveau de désordre d’un réseau dynamique. Nous proposons deux algorithmes simples et efficaces qui exploitent la structure temporelle d’un réseau dynamique pour délivrer les messages avec un bon compromis entre l’usage des ressources et les performances. Les résultats de simulations et analytiques montrent que ce type d’algorithme est plus performant que les approches classiques. Nous mettons également en évidence aussi la structure de réseau pour laquelle ce type d’algorithme atteint ses performances optimum. Basé sur ce résultat théorique nous proposons un nouveau protocole de routage efficace pour les réseaux opportunistes centré sur le contenu. Dans ce protocole, les noeuds maintiennent, via leurs contacts opportunistes, une fonction d’utilité qui résume leur proximité spatio-temporelle par rapport aux autres noeuds. En conséquence, router dans un tel contexte se résume à suivre le gradient de plus grande pente conduisant vers le noeud destination. Cette propriété induit un algorithme de routage simple et efficace qui peut être utilisé aussi bien dans un contexte d’adressage IP que de réseau centré sur les contenus. Les résultats de simulation montrent que ce protocole superforme les protocoles de routage classiques déjà définis pour les réseaux opportunistes. La dernière contribution de cette thèse consiste à mettre en évidence une application potentielle des réseaux dynamiques dans le contexte du « mobile cloud computing ». / This thesis contributes to the modeling, understanding and efficient communication in dynamic networks populating the periphery of the Internet. By dynamic networks, we refer to networks that can be modeled by dynamic graphs in which nodes and links change temporally. In the first part of the thesis, we propose a new mobility model - STEPS - which captures a wide spectrum of human mobility behavior. STEPS implements two fundamental principles of human mobility: preferential attachment and attractor. We show that this simple parametric model is able to capture the salient statistical properties of human mobility such as the distribution of inter-contact/contact time. In the second part, using STEPS, we analyze the fundamental behavioral and structural properties of opportunistic networks. We redefine in the context of dynamic networks the concept of small world structure and show how such a structure can emerge. In particular, we show that highly dynamic nodes can play the role of bridges between disconnected components, helping to significantly reduce the length of network path and contribute to the emergence of small-world phenomenon in dynamic networks. We propose a way to model this phenomenon in STEPS. From a regular dynamic network in which nodes limit their mobility to their respective preferential areas. We rewire this network by gradually injecting highly nomadic nodes moving between different areas. We show that when the ratio of such nomadic nodes is around 10%, the network has small world structure with a high degree of clustering and a low characteristic path length. The third contribution of this thesis is the study of the impact of disorder and contact irregularity on the communication capacity of a dynamic network. We analyze the degree of disorder of real opportunistic networks and show that if used correctly, it can significantly improve routing performances. We then introduce a model to capture the degree of disorder in a dynamic network. We propose two simple and efficient algorithms that exploit the temporal structure of a dynamic network to deliver messages with a good tradeoff between resource usage and performance. The simulation and analytical results show that this type of algorithm is more efficient than conventional approaches. We also highlight also the network structure for which this type of algorithm achieves its optimum performance. Based on this theoretical result, we propose a new efficient routing protocol for content centric opportunistic networks. In this protocol, nodes maintain, through their opportunistic contacts, an utility function that summarizes their spatio-temporal proximity to other nodes. As a result, routing in this context consists in following the steepest slopes of the gradient field leading to the destination node. This property leads to a simple and effective algorithm routing that can be used both in the context of IP networks and content centric networks. The simulation results show that this protocol outperforms traditional routing protocols already defined for opportunistic networks. The last contribution of this thesis is to highlight the potential application of dynamic networks in the context of "mobile cloud computing." Using the particle optimization techniques, we show that mobility can significantly increase the processing capacity of dynamic networks. In addition, we show that the dynamic structure of the network has a strong impact on its processing capacity.

Cette thèse apporte des contributions à la modélisation, compréhension ainsi qu'à la communication efficace d'information dans les réseaux dynamiques peuplant la périphérie de l'Internet. Par réseaux dynamiques, nous signifions les réseaux pouvant être modélisés par des graphes dynamiques dans lesquels noeuds et liens évoluent temporellement. Dans la première partie de la thèse, nous proposons un nouveau modèle de mobilité - STEPS - qui permet de capturer un large spectre de comportement de mobilité humains. STEPS mets en oeuvre deux principes fondamentaux de la mobilité humaine : l'attachement préférentiel à une zone de prédilection et l'attraction vers une zone de prédilection. Nous proposons une modélisation markovienne de ce modèle de mobilité. Nous montrons que ce simple modèle paramétrique est capable de capturer les caractéristiques statistiques saillantes de la mobilité humaine comme la distribution des temps d'inter-contacts et de contacts. Dans la deuxième partie, en utilisant STEPS, nous analysons les propriétés comportementales et structurelles fondamentales des réseaux opportunistes. Nous redéfinissons dans le contexte des réseaux dynamiques la notion de structure petit monde et montrons comment une telle structure peut émerger. En particulier, nous montrons que les noeuds fortement dynamiques peuvent jouer le rôle de ponts entre les composants déconnectés, aident à réduire significativement la longueur du chemin caractéristique du réseau et contribuent à l'émergence du phénomène petit-monde dans les réseaux dynamiques. Nous proposons une façon de modéliser ce phénomène sous STEPS. À partir d'un réseau dynamique régulier dans lequel les noeuds limitent leur mobilité à leurs zones préférentielles respectives. Nous recablons ce réseau en injectant progressivement des noeuds nomades se déplaçant entre plusieurs zones. Nous montrons que le pourcentage de tels nœuds nomades est de 10%, le réseau possède une structure petit monde avec un fort taux de clusterisation et un faible longueur du chemin caractéristique. La troisième contribution de cette thèse porte sur l'étude de l'impact du désordre et de l'irrégularité des contacts sur la capacité de communication d'un réseau dynamique. Nous analysons le degré de désordre de réseaux opportunistes réels et montrons que si exploité correctement, celui-ci peut améliorer significativement les performances du routage. Nous introduisons ensuite un modèle permettant de capturer le niveau de désordre d'un réseau dynamique. Nous proposons deux algorithmes simples et efficaces qui exploitent la structure temporelle d'un réseau dynamique pour délivrer les messages avec un bon compromis entre l'usage des ressources et les performances. Les résultats de simulations et analytiques montrent que ce type d'algorithme est plus performant que les approches classiques. Nous mettons également en évidence aussi la structure de réseau pour laquelle ce type d'algorithme atteint ses performances optimum. Basé sur ce résultat théorique nous proposons un nouveau protocole de routage efficace pour les réseaux opportunistes centré sur le contenu. Dans ce protocole, les noeuds maintiennent, via leurs contacts opportunistes, une fonction d'utilité qui résume leur proximité spatio-temporelle par rapport aux autres noeuds. En conséquence, router dans un tel contexte se résume à suivre le gradient de plus grande pente conduisant vers le noeud destination. Cette propriété induit un algorithme de routage simple et efficace qui peut être utilisé aussi bien dans un contexte d'adressage IP que de réseau centré sur les contenus. Les résultats de simulation montrent que ce protocole superforme les protocoles de routage classiques déjà définis pour les réseaux opportunistes. La dernière contribution de cette thèse consiste à mettre en évidence une application potentielle des réseaux dynamiques dans le contexte du " mobile cloud computing ". En utilisant les techniques d'optimisation particulaires, nous montrons que la mobilité peut augmenter considérablement la capacité de calcul des réseaux dynamiques. De plus, nous montrons que la structure dynamique du réseau a un fort impact sur sa capacité de calcul.

Cloud gaming over mobile networks enables players to play high-resource consuming games on low-end devices with various intrinsic restrictions such as limited battery lifetime and computational capacity. For mobile cloud gaming(MCG), the remaining battery level on the device is one of the critical factors that affect the sensitivity of user satisfaction. Thus, an android application is developed to measure the energy consumption of mobile devices that measure the power consumption of the device such that the obtained values correspond with the specific network conditions and users. The collected values are studied to identify if the energy consumption of the device is impacted by the network degradation that might occur during MCG in cellular networks. Results demonstrate that the energy consumption is at its highest when packet loss is 45% at 2ms RoundTrip Time (RTT) delay. Moreover, a qualitative study on the perceived Quality of Experience (QoE) of MCG over mobile networks is conducted and its impact on the energy consumption of the device is investigated where 31 users play a cloud-based First Person Shooter (FPS) for approximately 2 hours each. The results demonstrate the existence of the relationship between energy consumption and perceived QoE whereas negates the hypothesis of the existence of the relationship between QoE and CPU resources. In addition, to make comparisons of energy consumption of MCG with online mobile gaming (OMG), another test is carried out where each user plays another non-cloud-based FPS game and it is found that MCG is more efficient than OMG under the least energy-consuming network condition (2ms RTT delay) by 33.3% and the most energy consuming network condition (45% packet loss at 2ms RTT) by 32.7% in 4G cellular network.