Learn skills required for the MPLS portion of the Cisco CCIP certification. These skills are tested in the CCIP 642-611 -- Implementing Cisco MPLS Exam.
Make sure you understand the history of MPLS. Why was it desired?
MPLS is the combination of ATM capabilities over an IP backbone. IP has run over ATM circuits for quite some time, however, scalability has always been an issue with service providers as they provided redundant circuits between the routers that served customers for Internet traffic and also provided a mesh of circuits for the ATM backbone. ATM provided traffic engineering capabilities and quality of service. This was termed the overlay model, as IP traffic was overlayed on a mesh of ATM circuits.
Understand what is meant by an "overlay" model and why the meshing issue causes scalability problems.
The more routers the more ATM circuits were required to provide the mesh. The issue with this is that for every router added, there needed to be a mesh of circuits built to all the other routers. This can be demonstrated by the formula n(n-1)/2 where n is the number of routers. The formula derives the number of circuits required to provide a full mesh. If you have four routers, you would need six circuits to provide a full mesh. 100 routers would require 4500 circuits.
Be sure to understand the label stack.
The MPLS label is the foundation for label switching. The MPLS label has four octets, or 32 bits, that make up four fields. The fields are:
- The label field: Composed of 20 bits which allows for the creation of over one million labels (220 - 1, or 1,048,575 to be precise).
- The EXP field: Maps directly to IP Precedence TOS bits to provide class of service (COS) markings for an MPLS label. This field is three bits in length.
- The S field: Used for stacking labels. This is important and is used to indicate that last label in the label stack. The S field is one bit in length.
- The TTL field: Used to decrement the time-to-live counter. It is eight bits in length.
Each of these fields plays an important role in the delivery of MPLS technologies such as the creation and forwarding of traffic along a label-switched path, QoS guarantees and transport of one carrier's MPLS over another's backbone.
Be familiar with the field names and their purpose and length.
The MPLS label stack is inserted into ordinary packets between the IP header and the Layer 2 header (frame relay, Ethernet or ATM). This allows routers to switch the packet based on the MPLS label rather than the Layer 3 or Layer 2 information. This is why the MPLS label stack is sometimes referred to as the "shim header," as it is shimmed in between the Layer 2 and Layer 3 headers.
The MPLS label stack is just like an IP packet in that the router will make forwarding decisions based on the MPLS label found in the packets. The routers will use MPLS labels exclusively to forward traffic over an MPLS backbone; therefore, an understanding of the contents of the label is mandatory to understanding MPLS.
The next article will focus on how the router makes decisions on forwarding of packets containing MPLS labels. We will revisit the binding of labels to an IP route and discuss the control and forwarding plane of an MPLS router. This will provide the concepts necessary to understand how routers build the virtual circuits or label switched paths that the labels are forwarded across.
About the author:
Robbie Harrell (CCIE#3873) is the National Practice Lead for Advanced Infrastructure Solutions for SBC Communications. He has over 10 years of experience providing strategic, business, and technical consulting services to clients. Robbie resides in Atlanta, and is a graduate of Clemson University. His background includes positions as a Principal Architect at International Network Services, Lucent, Frontway and Callisma.