Tutorial on MPLS network, MPLS label format and MPLS frame
MPLS stands for Multi-Protocol Label Switching. Basically it uses label switching type as described below and it uses protocol independent transport mechanism. It is a transport mechanism which directs data from one node to the other node with the use of labels instead of IP addresses. Hence MPLS avoids use of long IP addresses as well as complex routing tables used for routing. The standard specifications are managed by IETF in order to develop a common standard.
Label switching mechanism is used in MPLS. The MPLS frame uses various different data link layer frames such as ATM, frame relay, T1/E1, synchronous optical network etc.
Why MPLS ?
As we know switching is used to connect two circuits in order to exchange information.
The switching is divided into two main types viz. circuit switching and packet switching.
Circuit switching is connection oriented routing and dedicated path is used during entire transmission of
Packet switching is connectionless routing and does not need dedicated path. In packet switching entire message is first divided into packets. These packets are addressed and numbered. These packets are transmitted one by one. These packets follow different routes from the source to the destination. At the destination these re-ordered packets are sequenced/ordered and information is retrieved. Refer Circuit Switching vs Packet Switching for more information.
As mentioned packet switching has some advantages over circuit switching. It also has several drawbacks.
The drawbacks of packet switching is overcome by label switching known as MPLS.
Following are the characteristics of label switching technique.
• Label switching converts connection-less routing to connection oriented routing.
• Label is attached to data packets according to class and type of services similar to categories/priorities in circuit switching.
• For routing across intermediate routers, apropriate labels are used.
MPLS Network Architecture and How does it work
As part of MPLS tutorial let us understand MPLS network and its working operation. Figure-1 depicts MPLS network architecture. As shown typically architecture of MPLS composed of CE, LER and LSR as described below. As shown the data path between the routers through which packets travel is known as LSP(Label Switched Path).
Before we understand MPLS architecture let us understand its terminologies.
• CE ( Customer Edge ) : It structures the customer message into IP packets. It later sends the same to entry node of MPLS domain. When it receives the packets from MPLS domain, it sends the packets to the network layer of its own after removing the IP address.
• LER ( Label Edge Router) : This functions as gateway of MPLS domain and sits at the edge of MPLS domain. Ingress LER receives IP packet from CE and it assigns the appropriate label to it. After assigning the label, it sends labelled packet towards the next hop through the LSP. Assignment of label is known as label binding. LER can also function as egress router. It receives labelled IP packets from previous router, removes the label and re-routes it to the destined CE.
• LSR (Label Switching Router ): This router type is used as transit switch in MPLS cloud. It basically does the job of MPLS forwarding. It receives the packets embedded with labels, analyzes it and routes them according to the labels mentioned in the LIB(Label Information Base) table. LIB is the database created both in the LER and LSR. It consists of incoming label and outgoing label. When LSR is routing the packets from incoming LSP to outgoing LSP, it removes the incoming label and assigns a new label to the same packet. This will provide security in MPLS network against any intruders. This process of changing label is also known as label swapping in MPLS network.
• LSP (Label Switched Path ) : Within MPLS domain, a dedicated path is setup prior to data transmission for a given packet to travel based on FEC. There are two types of LSPs viz. static and signalled. Static LSPs are configured manually on each of the LSR. No signalling protocol is used here.In order to establish static LSP, operator configures ingress LER, egress LER and transit LSRs. It also manually specify labels to each hops. Signalled LSPs are only configured at ingress LER. When a packet is assigned to signalled LSP, it will follow pre-established path from LSP's ingress LER to its egress LER counterpart.
• FEC (Forwarding Equivalence Class): The MPLS protocol groups packets having same characteristics into a common class. This class is known as FEC. The packets having same class are handled similarly on MPLS network.
• LFIB (Label Forwarding Information Base): It has same functionality as FIB (Forwarding Information Base) used on IP network.
When LSR entity receives a labeled packet. First it searches the LFIB database to obtain information in order to forward the packet. The information typically used for forwarding are label operation type, outgoing label value and next hop.
MPLS Label Format | MPLS Frame format
Figure-2 depicts Generic MPLS label format or MPLS frame format.
As shown a MPLS label is inserted between layer 2 header and layer 3 header of any packet.
It is four bytes in size and consists of following fields.
• Label: 20 bit size label value
• TC: 3 bit in size. It defines class of service. It is used for QoS.
• S (Stack): 1 bit in size. A label stack can comprise multiple labels. The label nearest to the Layer 2 header is called "top label". The label nearest to the Layer 3 header is called "bottom label". The S field is set to value 1 if the label is bottom label and set to 0 for all other label stack entries.
• TTL (Time to Live): 8 bit in size, It is used for routing loop prevention.
List of MPLS service providers
Following is the list of MPLS network service providers:
• Windstream Communications
• Interoute Communications Limited
• Level 3 Communications
• BT and TATA Docomo provides MPLS IP VPN service in INDIA.
The selection of MPLS in networking over other technologies depends on various factors, including the organization's specific requirements, budget and existing network infrastructure. As technology continues to evolve, newer options like Software Defined Wide Area Networks (SD-WAN) are also gaining prominence.
Networking related links
What is Router
What is Gateway
TCP-IP Packet format
Difference between OSI and TCPIP layers
Difference between TCP UDP
Packet Switching vs Message Switching