Cisco Multiservice Switching Networks

MPLS provides benefits that service providers desperately need in their networks, such as predictability, scalability, and manageability, all built into one network. The key to MPLS labels is that they tell a device not just where to send packets, but also how to send them. All the information needed is encoded in the label, including the destination prefix, the service class, the QoS, the level of privacy, the VPN, and so on. The network does not have to make a decision at every device. It's all precomputed in the LSP by means of the label. MPLS is transparent to both routers and switches because it is largely a switching intelligence. This is why MPLS is so scalable. It is how you deliver services across very large networks. You make service decisions once, at the edge, and the core automatically supports services without reprocessing packets at every stop. Because hop-by-hop routing decisions no longer have to be made, switched networks, such as carrier ATM backbones, can implement end-to-end Layer 3-type intelligence.

In addition, MPLS provides QoS capabilities that service providers need in order to offer differentiated services. Although MPLS might require modifications to existing equipment, it does not require extensive equipment replacement (no forklift upgrades). You can add this functionality by adding a new card, controller, and/or software to an existing ATM switch. Overall, MPLS defines an evolutionary networking paradigm that combines the operating principles of Layer 2 and Layer 3 technologies while preserving service providers' investment in IP routing technology at the network's edge and switching technology in the network's core.

MPLS VPN technology is one of the most important drivers for large MPLS deployments.

In addition, all the RAS features developed for carrier-class ATM switches are leveraged to provide high-availability LSRs.