Conventionally, transport networks were configured and redesigned manually. This process might be time consuming and in order to get the network operate smoothly careful network planning was required which might also include network downtime in some cases which was very critical. Reconfiguring services was also slow and painful because any mistakes could affect other services in the established network. It is obvious that due to more sophisticated transport networks, provision for control plane and hence traffic engineering will be easier.
As interest grew in providing efficient control plane to the transport networks, one choice was to develop a new set of protocols from scratch for all kinds of transport networks: one for WDM networks, one for TDM networks, and so on. The advantage of such approach would be that every control plane can be designed to be very economical for the target network. But at the same time the obvious disadvantage to individually configured control planes is that the huge quantity of effort required developing the numerous new sets of signaling, routing, and traffic engineering protocols and applications. Multilayer architecture having such transport networks would be very slow to scale and fairly cost-ineffective. Such multilayer architectures typically suffer from the lowest common denominator effect where any one layer can limit the scalability of the lowest common denominator effect where any one layer can limit the scalability of the entire network. End-to-end provisioning on such heterogeneous networks, every with its own separate management plane, would be a formidable task.
OVERVIEW AND BACKGROUND OF MPLS.
Fortunately, MPLS techniques prove to be a technology with similar switching notions that could be employed on heterogeneous networks to solve the traffic engineering problems that need to be addressed for all kinds of transport networks. So the MP Lambda S work was broadened to cover every existing switching technology such as fiber switching, TDM, layer 2 switching, and packet/frame/cell switching technologies.