MPLS Architecture Overview

MPLS Architecture Overview V1.1

Course Objectives • MPLS overview • MPLS Concepts • LSRs and labels • Label assignment and distribution • Label Switch Paths • Loops and TTL • LDP overview • Day in the Life of a Packet

MPLS Concepts • MPLS is a layer 2+ switching • Developed to integrate IP and ATM • MPLS forwarding is done in the same way as in ATM switches • Packet forwarding is done based on Labels

MPLS Origin • MPLS is developed by integrating IP switching protocols of different manufactures • IP Switching (Ipsilon) • Tag Switching (Cisco) • Aggregate RouteBased IP Switching (IBM) • Cell Switch Router (Toshiba) • IP Navigator (Ascend)

MPLS Origin • MPLS is developed by the combination of ATM and IP at 1990s. In 1996 , Ipsilon proposed IP switching protocol which solve the problem of better supporting IP on ATM switch, this makes an ATM switch a router and at the same time having the high performance of an ATM switch，breaking the performance bottleneck on the router (can not achieve wire speed on POS3,POS12)

When OC48,OC192 interface achieves wire speed, MPLS research switched to these application：TE (Traffic engineering), VPN. MPLS TE becomes an important method to manage network traffic、reduce congestion and ensure QOS on IP network. VPN is regarded as an important method to provide value-added service and explore new service by carriers. Disadvantages：Change the current network topology, increase the network complexity, VPN security factors. MPLS History

Multi-protocol label switch Label is at the layer 2.5, between link layer and network layer. This technology have the advantage of layer 2 switching and layer 3 routing. The successful point of MPLS technology is that it introduced connection-oriented mechanism into IP network; Core idea of MPLS is：routing on the edge and switching on the core part Establishing of LSP（label switch path）：LSP is established by topology not by data. The establishing of LSP can be done by LDP，CR-LDP，RSVP protocols. MPLS Overview

MPLS Developing Motion • Current core network is made up of router and ATM switch • Core network based on MPLS is superior than core network based on router and ATM • This is the developing motion of MPLS.

Comparison between MPLS and routing • MPLS simplifies the process of packet forwarding • MPLS supports valid explicit routing • MPLS can implement TE best • MPLS can support QoS routing • Mapping from IP packet to FEC • MPLS support dividing network by function • MPLS implements single forwarding rule no matter how many levels of Qos are there

Advantage of MPLS • Comparison of label switching and hop-by-hop forwarding based on destination IP address： • MPLS is implemented on switch, thus taking full advantage of high speed of switch. Traditional switch can search label and replace them, but can not analyze network header or at least not fast enough to analyze it, so it is difficult to implement forwarding based on network header information on traditional switch • When packet comes into MPLS network, ingress router will decide which FEC it belongs to, it can use any information about the packet not only the network layer information.（for example the port information）

Advantage of MPLS (Continued.) • MPLS can assign different labels to packet coming from different routers, thus making it easy to implement routing based on ingress routers. Because ingress router id can not be sent with the packet , so traditional routing patterns do not support this function. • Sometimes it is reasonable to designate a predefined route for the packets than hop-by-hop routing.

Unlike IP, classification/label can be based on: Destination Unicast address Traffic Engineering VPN QoS FEC: Forwarding Equivalence Class A FEC can represent a: Destination address prefix, VPN, Traffic Engineering tunnel, Class of Service. MPLS Concepts

Key words in MPLS • Label • FEC (Forwarding equivalence class) • LSR (Label switching router) • LER (Label edge router) • LSP (Label switching path) • LDP (Label distribution path)

Label is a key point in MPLS, it is a short fixed-length and locally used identifier which is used to identify FEC Label processing is done by high speed ASIC chip thus making the delay of packet processing and queuing decrease to a great extent Why fixed-length label? Balancing between forwarding efficiency and switching performance. Although fixed-length making the forwarding efficiency low but it can highly improve switching performance. Label is locally used identifier. It is only meaningful between the sending out port of upstream router and receiving port of downstream router. Label

MPLS label is an integer ranged between 0 to 1048575,it is used to identify specific FEC. This label is encapsulated in layer 2 header. Label Format

Two or more MPLS labels, encapsulated after link layer header and before network header. Top label in the stack appear earliest. Network layer header follow the last label in the stack Label stack

Label stack • Forwarding of packets is based on the top label in the stack，when LSR receives a packet，checks the top label and decides the next hop • The operation about the label stack： • replace：Use a new label to replace the top label in the stack • Pop (delete)：pop the top label in the label stack and delete it • push (add)：replace the top label and add a new label into the stack

FEC • Forwarding Equivalence Class (FEC)：A group of packets that have common attributes. These packets will be forwarded in the same way by LSR in the MPLS network，just because they are forwarded in the same way they are equivalence.

Ingress FECp+q FECp LERa LERe LERb LERd LERf LERc Egress LSP LSRy FECq LSRx LSRz MPLS Domain FEC • FEC – packets forwarded in the same way • Same destination prefix unicast packet • Same destination address multicast packet • Same Qos packets

LSR –responsible for establishing LSP for FEC LER –responsible for FEC classification, TE，begin the process of establishing LSP, IP packet forwarding LSP –used for IP packet forwarding LDP –responsible for assigning labels Ingress FECp+q FECp LERc LERb LERa LERd LERf LERe Egress LSP LSRy FECq LSRx LSRz MPLS Domain MPLS key points

MPLS key points • Label switch router (LSR)：one node in MPLS network, it is located in the middle of MPLS network, it runs MPLS control protocol and layer 3 routing protocol, its responsibility lies in: • exchanging routing information with other LSR to form route table, • implementing the mapping from FEC to IP packets，redistributing label binding information, establishing label forwarding table and maintaining it.

MPLS key points • Label edge router (LER)：Responsible for connecting MPLS domain and non-MPLS domain. Implementing the function of classifying service、assigning labels、taking off labels etc. LER is the key device in implementing MPLS.

MPLS Key Points • Label switch path (LSP)：The forwarding path made up of the LSRs along the way to forward packets which belong to the same FEC. • Label distributing protocol (LDP)：Responsible for controlling the label binding information exchanging process between LSRs，LSR establish and maintain the LIB (Label information base) according to the binding information between label and FEC.

MPLS Key points • Upstream router • Downstream router

LSRs and Labels • LSR: Label Switch Router • Edge-LSR: LSRs that do label imposition and disposition • ATM-LSR: An ATM switch with Label Switch Controller

An IP routing protocol is used within the routing domain (e.g.:OSPF, i-ISIS) A label distribution protocol is used to distribute address/label mappings between adjacent neighbors The ingress LSR receives IP packets, performs packet classification, assign a label, and forward the labelled packet into the MPLS network Core LSRs switch packets/cells based on the label value The egress LSR removes the label before forwarding the IP packet outside the MPLS network LSRs and Labels IGP domain with a label distribution protocol

PPP Header(Packet over SONET/SDH) PPP Header Shim Header Layer 3 Header Ethernet Ethernet Hdr Shim Header Layer 3 Header Label Label Frame Relay FR Hdr Shim Header Layer 3 Header GFC VPI VCI PTI CLP HEC DATA ATM Cell Header Subsequent cells GFC VPI VCI PTI CLP HEC DATA LSRs and Labels

Label Assignment and Distribution • Labels have link-local significance • Each LSR binds his own label mappings • Each LSR assign labels to his FECs • Labels are assigned and exchanged between adjacent neighboring LSR • Applications may require non-adjacent neighbors

Rtr-C is the downstream neighbor of Rtr-B for destination 171.68.10/24 Rtr-B is the downstream neighbor of Rtr-A for destination 171.68.10/24 LSRs know their downstream neighbors through the IP routing protocol Next-hop address is the downstream neighbor Label Assignment and Distribution Upstream and Downstream LSRs 171.68.40/24 171.68.10/24 Rtr-A Rtr-B Rtr-C

LSRs distribute labels to the upstream neighbors Use label 30 for destination 171.68.10/24 Use label 40 for destination 171.68.10/24 In I/F In I/F In I/F In Lab In Lab In Lab Address Prefix Address Prefix Address Prefix Out I/F Out I/F Out I/F Out Lab Out Lab Out Lab 0 0 0 30 - 40 171.68.10 171.68.10 171.68.10 1 1 1 40 - 30 Next-Hop Next-Hop Next-Hop ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... Label Assignment and Distribution Unsolicited Downstream Distribution 171.68.40/24 171.68.10/24 Rtr-A Rtr-B Rtr-C IGP derived routes

Upstream LSRs request labels to downstream neighbors Downstream LSRs distribute labels upon request Use label 30 for destination 171.68.10/24 Use label 40 for destination 171.68.10/24 Request label for destination 171.68.10/24 Request label for destination 171.68.10/24 Label Assignment and Distribution On-Demand Downstream Distribution 171.68.10/24 171.68.40/24 Rtr-A Rtr-B Rtr-C

Label Assignment and Distribution • Label Retention Modes • Liberal retention mode • LSR retains labels from all neighbors • Improve convergence time, when next-hop is again available after IP convergence • Require more memory and label space • Conservative retention mode • LSR retains labels only from next-hops neighbors • LSR discards all labels for FECs without next-hop • Free memory and label space

Label Distribution Modes Independent LSP control LSR binds a Label to a FEC independently, whether or not the LSR has received a Label the next-hop for the FEC The LSR then advertises the Label to its neighbor Ordered LSP control LSR only binds and advertise a label for a particular FEC if: it is the egress LSR for that FEC or it has already received a label binding from its next-hop Label Assignment and Distribution

Several protocols for label exchange LDP Maps unicast IP destinations into labels RSVP, CR-LDP Used in traffic engineering BGP External labels (VPN) PIM For multicast states label mapping Label Assignment and Distribution

LSPs are derived from IGP routing information LSPs may diverge from IGP shortest path LSP tunnels (explicit routing) with TE LSPs are unidirectional Return traffic takes another LSP Label Switch Path (LSP) IGP domain with a label distribution protocol IGP domain with a label distribution protocol LSP follows IGP shortest path LSP diverges from IGP shortest path

LSP establishing process • The establishing of LSP in MPLS network including these three steps: • Every node run routing protocols such as BGP、OSPF、IS-IS to form its own route table • According to the route table, every node establish label information base under the control of LDP • From the ingress LSR 、middle LSR and egress LSR, the ingress lable and outgress lable together form a LSP.

3 47.1 1 2 1 3 2 1 47.2 3 47.3 2 1st step：form of route table • Dynamic routing protocols help each router form route table.

47.1 1 IP 47.1.1.1 2 IP 47.1.1.1 1 3 2 IP 47.1.1.1 1 47.2 3 47.3 2 IP 47.1.1.1 Traditional ‘hop by hop’ forwarding

Request: 47.1 Request: 47.1 Mapping: 0.40 Mapping: 0.50 2nd step：form of LIB 1 47.1 3 2 3 1 1 2 47.3 3 47.2 2

IP 47.1.1.1 IP 47.1.1.1 3rd step：form of LSP 1 47.1 3 3 2 1 1 2 47.3 3 47.2 2

Routing method in MPLS • hop-by-hop routing • This method allows each node to select the next hop for each FEC independently • This kind of routing method is commonly used in IP network

Routing method in MPLS • Explicit Routing • In this kind of routing method, each LSR can not select next hop independently，on the contrary，path selection is done under the network management policy, for example, the ingress or egress LSR define the way of the LSP. • When the ingress of egress LSR define every hop along the LSP , we call it “strict explicit routing”，if it only define part of the nodes along the way，we call it “loose explicit routing” • “strict explicit routing” is also called “source routing” in IP network，but compared to IP source routing, strict explicit routing has higher efficiency.

47.1 1 IP 47.1.1.1 2 IP 47.1.1.1 1 3 2 IP 47.1.1.1 1 47.2 3 47.3 IP 47.1.1.1 2 hop-by-hop routing

Route={A,B,C} #972 #14 #216 #14 B #972 C A #462 Explicit Routing

IP 47.1.1.1 1 47.1 3 3 2 1 1 2 47.3 3 47.2 2 IP 47.1.1.1 ER-LSP

Label Switch Path (LSP) Penultimate Hop Popping • The label at the top of the stack is removed (popped) by the upstream neighbor of the egress LSR • The egress LSR requests the “popping” through the label distribution protocol • Egress LSR advertises implicit-null label • The egress LSR will not have to do a lookup and remove itself the label • One lookup is saved in the egress LSR

MPLS Architecture Overview