Y. Shi et al. (Eds.): ICCS 2007, Part I, LNCS 4487, pp. 628–631, 2007. © Springer-Verlag Berlin Heidelberg 2007 Traffic Routing Through Off-Line LSP Creation Srecko Krile and Djuro Kuzumilovic University of Dubrovnik, Department of Electrical Engineering and Computing, Cira Carica 4, 20000 Dubrovnik, Croatia srecko.krile@unidu.hr Abstract. In the context of dynamic bandwidth allocation the QoS path provisioning for coexisted and aggregated traffic could be very important element of resource management. As we know all traffic in DiffServ/MPLS network is distributed among LSPs (Label Switching Path). Packets are classified in FEC (Forwarding Equivalence Class) and can be routed in relation to CoS (Class of Service). In the process of resource management we are looking for optimal LSP, taking care of concurrent flows traversing the network simultaneously. For better load-balancing purposes and congestion avoidance the LSP creation can be done off-line, possible during negotiation process. The main difference from well known routing techniques is that optimal LSP need not to be necessarily the shortest path solution as it is in the case of typical on- line routing (e.g. with OSPF protocol). Keywords: on-demand resource allocation, dynamical bandwidth management, end-to-end QoS routing, SLA in DiffServ/MPLS networks. 1 Introduction Using dynamic service negotiation approach for SLA (Service Level Agreement) the problem of QoS path provisioning has to be in firm correlation with bandwidth management; see [1]. Aggregated flow consisted of numerous LSPs (Label Switching Path) is coming to LSR (Label Switched Router) and has to be routed to egress router. All traffic, traversing simultaneously a DiffServ/MPLS domain, is distributed among LSPs as the resultant of the routing protocol. Some of LSPs generally traverse through the same path across the network, so they coexist on the same link with congestion probability; see [2]. In that sense the network operator has to find the optimal LSP for each aggregated flow without any possible congestion in the network; see [6]. In the context of on-demand resource allocation the load balancing technique is necessary. The main condition is: the sufficient network resources must be available for the priority traffic at any moment. In this paper such off-line routing technique is proposed. Proposed heuristic algorithm helps in bottleneck detection on the path and maintains high network resource utilization efficiency. A brief explanation: During SLA negotiation the RM (Resource Manager) module can apply any shortest path-based routing algorithm (e.g. OSPF - Open Shortest Path First) to generate the initial LSP. At the next step all simultaneous flows caused by former contracted SLAs have to be taken in calculation, too (correlation with other LSPs); see [4]. The BB (Bandwidth Broker) will therefore check if there are enough resources to satisfy the requested CoS in the specified path. With such congestion control/expansion algorithm the RM can predict sufficient link resources to satisfy all