The paper proposes a resource efficient solution for Network Coding (NC) based congestion control consisting in identification around the congested links of multiple butterfly topologies by using the Discrete Lagrange Multiplier optimization algorithm. The instantiation of the butterfly topologies is based on the resource management capabilities foreseen for the entities of the future networks. The congestion control issue is tackled by separate encoding of appropriately selected pairs of data flows passing through the congested link. By optimal selection of the flows to be coded it is achieved the goal to minimize the additional network resources required by the NC operations. The encoding is realized with an XOR-based algorithm adapted for unequal bit rate data flows, and the experimental performances are reported here. Due to its efficient usage of the network resources and high degree of scalability, the congestion control solution proposed is suitable for large bit rate transmissions, like video streaming.