Multiprotocol Label Switching (MPLS) forwards traffic through a provider network using short labels rather than repeated IP routing lookups at every hop. It is widely used to deliver scalable Layer 3 VPN, Layer 2 VPN, traffic engineering, and QoS services.
How MPLS Works
At the ingress edge, a label edge router (LER) classifies an incoming packet into a forwarding equivalence class (FEC) and pushes one or more labels. Core label switch routers (LSRs) swap the top label according to their label forwarding information base. At the egress edge, the label is removed and the packet is forwarded normally.
MPLS Label Structure
- Label: 20-bit forwarding value.
- Traffic Class: 3 bits used for QoS-related handling.
- Bottom of Stack: indicates the final label in a stack.
- TTL: supports loop prevention and troubleshooting.
Key MPLS Terms
- LER: edge router that pushes or removes labels.
- LSR: core router that forwards based on labels.
- LSP: the label-switched path followed through the MPLS network.
- FEC: a group of packets forwarded in the same way.
Label Distribution
Labels are commonly distributed by LDP, RSVP-TE, Segment Routing, or BGP-based MPLS services, depending on the design. Interior routing still provides reachability between provider routers.
Common MPLS Uses
- Layer 3 VPNs for separated customer routing.
- Layer 2 VPNs for Ethernet connectivity across a provider network.
- Traffic engineering and resilient path design.
- QoS-aware transport for business applications.
Troubleshooting MPLS
Check IGP reachability first, then verify label bindings, the LFIB, MPLS interfaces, and VPN/BGP configuration where used. Common validation commands on Cisco platforms include show mpls interfaces, show mpls ldp neighbor, and show mpls forwarding-table.
Continue Learning
For related networking fundamentals, read our WAN Technologies guide and IP Addressing guide. For structured Cisco learning materials, explore the Cisco Premium PDF collection.