The most of the research activities in network resource management have been devoted to providing bandwidth guarantees and preventing network congestion. It is understandable because the carriers wish to maximize the usage efficiency of their network infrastructure. Also, the rapid growth real-time network applications (e.g. VoIP) has made it imperative to consider the impact of end-to-end delay of traffic requirements on network resource provisioning. Very important element for efficient end-to-end QoS routing is good prediction of traffic demands in the phase of the negotiation process -SLA (Service Level Agreement) creation. In this paper algorithm for end-to-end congestion control is being developed. As explicit traffic engineering technique it provides the opportunity to intelligently tailor the route such that different parts of the network remain equally loaded. Such an approach avoids the creation of bottleneck links and maintains high network resource utilization efficiency. The problem is seen as an expansion problem of link capacities in given limits from a common source for given traffic demands.