Many computers and devices such as smart phones, laptops and tablet devices are now equipped with multiple network interfaces, enabling them to use multiple paths to access content over the network. If the resources could be used concurrently, end user experience can be greatly improved. The recent studies in MPTCP suggest that improved reliability, load balancing and mobility are feasible. The thesis presents a new multipath delay based algorithm, MPPERT (Multipath Probabilistic Early response TCP), which provides high throughput and efficient load balancing. In all-PERT environment, MPPERT suffers no packet loss and maintains much smaller queue sizes compared to existing MPTCP, making it suitable for real time data transfer. MP-PERT is suitable for incremental deployment in a heterogeneous environment. It also presents a parametrized approach to tune the amount of traffic shift off the congested path. Multipath approach is benefited from having multiple connections between end hosts. However, it is desired to keep the connection set minimal as increasing number of paths may not always provide significant increase in the performance. Moreover, higher number of paths unnecessarily increase computational requirement. Ideally, we should suppress paths with low throughputs and avoid paths with shared bottlenecks. In case of MPTCP, there is no efficient way to detect a common bottleneck between subflows. MPTCP applies a constraint of best single-path TCP throughput, to ensure fair share at a common bottleneck link. The best path throughput constraint along with traffic shift, from more congested to less congested paths, provide better opportunity for the competing flows to achieve higher throughput. However, the disadvantage is that even if there are no shared links, the same constraint would decrease the overall achievable throughput of a multipath flow. PERT, being a delay based TCP protocol, has continuous information about the state of the queue. This information is valuable in enabling MPPERT to detect subflows sharing a common bottleneck and obtain a smaller set of disjoint subflows. This information can even be used to switch from coupled (a set of subflows having interdependent increase/decrease of congestion windows) to uncoupled (independent increase/decrease of congestion windows) subflows, yielding higher throughput when best single-path TCP constraint is relaxed. The ns-2 simulations support MPPERT as a highly competitive multipath approach, suitable for real time data transfer, which is capable of offering higher throughput and improved reliability.

Au cours des vingt dernières années, le secteur des services est devenu le secteur le plus important en nombre d'actifs occupés dans l’économie mondiale, en particulier dans les pays développés. Par ailleurs, la concurrence et la coopération dans le secteur des services sont devenues de plus en plus populaires dans le contexte de la mondialisation économique. Comment collaborer avec un accord gagnant-gagnant apporte une source fertile de problèmes de management des opérations dans le domaine des services. Dans cette thèse, nous étudions des stratégies de collaboration dans des systèmes de services homogènes. Nous nous concentrons en particulier sur les stratégies de pooling des ressources de service.Dans les deux premières parties, nous étudions le problème de partage des coûts entre les fournisseurs de services indépendants avec des temps de service qui suivent une distribution générale et en tenant compte de l'abandon des clients. Nous modélisons à la fois chaque fournisseur de services et la coalition coopérative comme des files d'attente avec serveur unique, et spécialisons les stratégies de pooling avec les capacités de services fixes et modifiables. Dans la dernière partie, nous abordons le problème de pooling dans le cadre multiserveur pour évaluer la qualité de l'hypothèse "superserveur". Nous étudions numériquement l'impact de la variabilité de la durée de service et l'abandon des clients sur les jeux de mise en commun des ressources. Nous comparons aussi les partages des coûts entre le système de "super-serveur" et multiserveur. / In past twenty years, the service sector has emerged as the primary sector in the world economy, especially in developed countries. Competition and cooperation in service industries have become more and more popular in the context of economic globalization. How to operate the collaboration with a win-win agreement brings a fertile source of operations management issues in service science. In this thesis, we study collaborations between homogeneous service systems in terms of resource pooling strategies.In the first two parts, we investigate the cost-sharing problem among independent service providers with general service times and accounting for the customer abandonment. We model both the service provider and the cooperative coalition as single server queues, and specialize the capacity pooling strategies with the fixed and optimized service capacities.Finally, we address the service pooling problem in the multi-serverpooling setting to assess the quality of the "super-server" assumption.We numerically investigate the impact of service duration variability and customer abandonment on the pooling game. We compare between cost-sharing results of the two resource pooling concepts, with or without the "super-server" assumptions.