Internet applications such as web or ftp will be satisfactorily supported by a guaranteed minimum throughput service. We propose to build this service in a core-stateless network such as Differentiated Services with an Admission Control (AC) method based on end-to-end measurements, using the first packets of the flow. The whole scheme uses a small set of packet classes (with different discarding priorities), and is therefore simple and scalable. We evaluate the scheme through simulations in several scenarios using TCP flows. Finally, we propose an extension of the method aimed at improving its performance.

Abstract--This paper considers a problem for scheduling jobs on two identical parallel machines, the aim was to minimize two criteria in particular, makespan and total flow time. In order to solve this problem, two approaches were considered. A mechanism was proposed as an approach to solve this type of problem with a setting of a 2-player non-cooperative game, under the framework of a 2x2 non-sum zero matrix; each player looking after one of the criteria suggested in the scheduling problem. On the other hand, a Genetic Algorithm, known as Strength Pareto Evolutionary Algorithm (SPEA), was applied to the problem. The comparison between both approaches suggests a complementarity among rational agents approach models and machine enforced solution approaches. The resulting Pareto Front set of points were plotted and curves were compared, showing promising results for game theoretic applications to scheduling under multiple objectives.Keywords-identical parallel machines, makespan, total flow time, non-cooperative game, SPEA, Pareto Front.Resumen--Este articulo contempla el problema de programación de la producción en una configuracion de maquinas en paralelo con el objetivo de minimizar dos criterios en particular: el lapso y el tiempo total de flujo. En este problema en particular, el incremento de uno de estos objetivos resulta en la reduccion del otro, por lo que se propone su solucion bajo enfoques metaheuristicos. Dos tipos de algoritmos fueron considerados: uno basado en la teoria de juegos y el otro en los algoritmos genéticos. Para el primero se diseña un mecanismo de juego no cooperativo entre dos jugadores, en donde cada jugador busca optimizar cada criterio de programación de las máquinas. Para el segundo enfoque se implementa el algortimo genético SPEA, en donde se seleccionan aquellas soluciones dominantes en ambos objetivos. Resultados de ambos enfoques resultan en un Frente de Pareto, las cuales representan las soluciones dominantes para ambos objetivos. Estos resultados demuestran que ambos enfoques son complementarios: SPEA arroja resultados que cubren todo el frente de Pareto, mientras que el algoritmo de Juegos No Cooperativo indica la programación mas conveniente para cada agente en particular.

This paper proposes a traffic engineering scheme using different distribution trees to several multicast flows. We formulate this problem as one with Non Linear programming with discontinuous derivatives (DNLP). The aim is to combine into a single aggregated metric the following weighting objectives: the maximum link utilization, the hop count, the total bandwidth consumption and the total end-to-end delay. Our proposal solves the traffic split ratio for multiple trees. The proposed approach is applied in MPLS networks by allowing the establishment of explicit routes in multicast events. Furthermore, the results obtained with GAMS tools show that several objectives are decreased; the maximum link utilization is minimized. The main contribution of this paper is the optimization model and the formulation of the multi-objective function.