Abstract

We consider the problem of allocating bandwidth to competing flows in an MPLS network, subject to constraints on fairness, efficiency, and administrative complexity. The aggregate traffic between a source and a destination, called a flow, is mapped to label switched paths (LSPs) across the network. Each flow is assigned a preferred ('primary') LSP, but traffic may be sent to other ('secondary') LSPs. Within this context, we define objectives for traffic engineering, such as fairness, efficiency, and preferred flow assignment to the primary LSP of a flow ('Primary Path First', PPF). We propose a distributed, feedback-based multipath routing algorithm that attempts to apply additive-increase and multiplicative-decrease (AIMD) to implement our traffic engineering objectives. The new algorithm is referred to as multipath-AIMD. We use ns-2 simulations to illustrate the fairness criteria and PPF property of our multipath-AIMD scheme in an MPLS network. Document type: Part of book or chapter of book

Abstract

We analyse the possible consequences of activating iBGP multipath load sharing in a given domain (or AS), which allows for load balancing over multiple exit routers. It has been stated that interdomain routing loops may appear in this case. We show that under reasonable assumptions (which reflect commercial relationships between ASes) such routing loops cannot appear. Furthermore we show that even if theses assumptions are not met, routing loops can only be transient. Document type: Part of book or chapter of book

Abstract

The main contribution of this paper is a new online routing algorithm, called Dynamic Online Routing Algorithm (DORA), for dynamic setup of bandwidth guaranteed paths in MPLS networks. The goal of DORA is to accept as many network path setup requests as possible by carefully mapping paths with reserved bandwidth evenly across the network. The key operation behind DORA is to avoid routing over links that: 1) have high potential to be part of any other paths, and 2) have less residual bandwidth available. We compare DORA against other existing constraint-based routing algorithms based on two performance metrics: path setup rejection ratio and percentage of successful reroutes. Our result shows that DORA performs better than the other algorithms in both metrics. Document type: Part of book or chapter of book

Abstract

Intra-domain traffic engineering is essential for the operation of an Internet Service Provider. Demand-oblivious routing [2] promises excellent performance guarantee with changing and uncertain traffic demands. However, it is difficult to implement it. We investigate an efficient and deployable implementation of oblivious routing. We study its performance by both numerical experiments and simulation. The performance study shows that the multipath implementation achieves a close approximation to oblivious routing [2], especially when approximate knowledge of traffic is available. The study also shows its robustness under varying traffic demands, link failures and an adversary attack. Its performance is excellent even with a 100% error in traffic estimation. Document type: Part of book or chapter of book

Abstract

Traffic Engineering is the one of the methods for improving the Quality of Service in the network like Internet. Multipath routing is a mechanism for load balancing in which the total load is spatially distributed over several paths. It can reduce congestion probability given by routing into best one path in the current Internet. We focus on MPLS network, which is usable in the ISP network. LSPs of MPLS provide path controllability by explicit route in the connectionless IP network."br /"Multipath routing scheme is established with two steps, selection of multiple candidate paths and traffic splitting among these selected paths. We compare some recent multipath routing algorithms based on MPLS in the point of scalability and stability. Of them, we pick up WDP and LDM for our hybrid approach which combines cross-steps of different algorithms. PER algorithm is proposed in the improvement of LDM's traffic splitting step. Using LDM, WDP and PER, three hybrid QoS algorithms by each step combination are made. Their performances are proved by simulation that they, especially LBWDP (Load Balancing over Widest Disjoint Paths), are more effective than LDM or MATE. "br /"For better QoS guarantee, we integrate DiffServ with our traffic engineering technique (DiffServ-aware MPLS Traffic Engineering: DS-TE). PEMS (PEriodic Multi-Step algorithm in DS-TE network) is proposed as one DS-TE model to differentiate the quality of service according to the class. For PEMS algorithm, we propose new algorithm to select candidate paths by proper metric of each traffic class. PER algorithm distributes the traffic among its own candidate paths by dynamic metric of requested class. Simulation results using ns-2 showed that PEMS can balance the load less than LBWDP but EF and AF classes can be ensured the better quality of service than LBWDP.; Le routage multi-chemins est une technique qui permet l'équilibrage de la charge en multiplexant les flux vers une destination sur plusieurs chemins. Nous proposons une approche de routage multi-chemins qui, peut être schématisée en deux étapes : l'étape de choix des chemins candidats et l'étape de distribution du trafic sur un sous-ensemble de ces chemins. "br /"Dans ce travail, nous avons commencé par effectuer une étude comparative d'un point scalabilité et stabilité, de plusieurs algorithmes de routage multi-chemins, basés sur MPLS. Cela nous a permis de retenir WDP pour la sélection des chemins candidats et LDM pour la distribution des demandes de trafic reçues par un routeur entrant du réseau d'un FAI. Nous proposons dans ce travail PER, un algorithme qui est une amélioration de l'algorithme de distribution de LDM. Ces différents algorithmes nous ont permis de réaliser plusieurs algorithmes « hybrides » dont LBWDP (Load Balancing over Widest Disjoint Paths) qui par simulation a été prouvé comme étant un algorithme plus performant que des modèles comme LDM ou MATE. "br /"Pour une meilleure garantie de QdS nous avons cherché à intégrer la différentiation de service (DiffServ) avec notre technique d'ingénierie de trafic (DiffServ-aware MPLS Traffic Engineering : DS-TE). Nous proposons PEMS (PEriodic Multi-Step algorithm for DS-TE network) comme un modèle de DS-TE pour différencier la qualité du service selon la classe du trafic. Dans ce cadre, nous proposons un nouvel algorithme de sélection de chemins candidats en fonctions des critères préférentiels de chaque classe de trafic. L'utilisation de PER permet ensuite de distribuer en fonction de critères dynamiques les demandes reçues sur les meilleurs chemins de chaque classe. Par simulation à l'aide de ns-2, nous avons montré que PEMS répartie moins bien la charge que LBWDP mais que les classes EF et AF ont une qualité de service meilleure que dans le cas de LBWDP.