Multiprotocol Label Switching

-Simhadri

Copyright © 2013 Tech Mahindra. All rights reserved. 1

Overview
 Need for MPLS
 MPLS Basics
 Benefits
 Label Switched Path
 Label Distribution Protocol
 Hierarchy in MPLS
 Explicit Routing
 Loop Detection
 Traffic Engineering
 Constraint Based Routing
 Tag Switching
 IP Switching

Copyright © 2013 Tech Mahindra. All rights reserved. 2

Conventional IP Networks & Routing
 Client networks are connected to backbone via edge routers
LAN, PSTN, ADSL
–
 Data packets are routed based on IP address and other
information in the header
 Functional components
 Forwarding
– responsible for actual forwarding across a router
– consists of set of procedures to make forwarding decisions
 Control
– responsible for construction and maintenance of the forwarding table
– consists of routing protocols such as OSPF, BGP and PIM

Copyright © 2013 Tech Mahindra. All rights reserved. 3

 Need for Multiprotocol Label Switching
 Forwarding function of a conventional router

a capacity demanding procedure

constitutes a bottle neck with increase in line
speed
 MPLS simplifies forwarding function by taking a totally
different approach by introducing a connection oriented
mechanism inside the connectionless IP networks

Copyright © 2013 Tech Mahindra. All rights reserved. 4

Label Switching
 Decomposition of network layer routing into control and
forwarding components applicable
 Label switching forwarding component algorithm uses
 forwarding table
 label carried in the packet
 What is a Label ?
 Short fixed length entity

Copyright © 2013 Tech Mahindra. All rights reserved. 5

MPLS Basics
 A Label Switched Path (LSP) is set up for each route
 A LSP for a particular packet P is a sequence of routers,
<R1,R2………..Rn>
for all i, 1< i < n: Ri transmits P to R[i+1] by means
of a label
 Edge routers
 analyze the IP header to decide which LSP to use
 add a corresponding local Label Switched Path Identifier,
in the form of a label
 forward the packet to the next hop

Copyright © 2013 Tech Mahindra. All rights reserved. 6

MPLS Basics contd..
 Subsequent nodes

just forward the packet along the LSP

simplify the forwarding function greatly

increase performance and scalability dramatically
 New advanced functionality for QoS, differentiated services
can be introduced in the edge routers
 Backbone can focus on capacity and performance
 Routing information obtained using a common intra domain
routing protocol such as OSPF

Copyright © 2013 Tech Mahindra. All rights reserved. 7

Basic Model for MPLS Network

Internet

LER

LER IP
LSR
LSR

MPLS LSR
LSR MPLS

LER IP

LSR = Label Switched Router

LER = Label Edge Router

Copyright © 2013 Tech Mahindra. All rights reserved. 8

MPLS Operation

Intf Label Intf Label Intf Label Intf

In In Out Out In In Out
3 50 1 40 3 40 1

FEC Intf Label 1

3 3
Out Out 1
a 1 50 2
40
1 2
50
3
2

9
Copyright © 2013 Tech Mahindra. All rights reserved. 9
MPLS Benefits
Comparing MPLS with existing IP core and IP/ATM
technologies, MPLS has many advantages and benefits:
 The performance characteristics of layer 2 networks
 The connectivity and network services of layer 3
networks
 Improves the price/performance of network layer
routing
 Improved scalability
 Improves the possibilities for traffic engineering
 Supports the delivery of services with QoS guarantees
 Avoids need for coordination of IP and ATM address
allocation and routing information
10
Copyright © 2013 Tech Mahindra. All rights reserved. 10
 Necessity of L3 Forwarding
 For security

To allow packet filtering at firewalls

Requires examination of packet contents, including the
IP header
 For forwarding at the initial router - used when hosts don’t
do MPLS
 For Scaling
 Forward on a finer granularity than the labels can
provide

11
Copyright © 2013 Tech Mahindra. All rights reserved. 11
 Carrying a Label
 Certain link layer technologies can carry label as a part
of their link layer header
 e.g ATM & Frame Relay
 Link layers that do not support labels in their header
carry them in a “shim” label header

Link layer “Shim” label Network Network

header header layer header layer data

12
Copyright © 2013 Tech Mahindra. All rights reserved. 12
MPLS Header Format

 Label: 20-bit label value

 Exp: experimental use
– Can indicate class of service
 S: bottom of stack indicator
– 1 for the bottom label, 0 otherwise
 TTL: time to live

20 3 1 8

Label Exp S TTL

13
Copyright © 2013 Tech Mahindra. All rights reserved. 13
 Establishing Label Switched Path
 LSPs are generated and maintained in a distributed
fashion
 Each LSR negotiates a label for each Forwarding
Equivalence Class (FEC) with its upstream and
downstream neighbors using a distribution method
 Label Information Base (LIB) - Result of negotiation

14
Copyright © 2013 Tech Mahindra. All rights reserved. 14
 LDP - Terminology
 Label Distribution Protocol (LDP)
– set of procedures by which LSRs establish LSPs
– mapping between network-layer routing information
directly to data-link layer switched paths
 LDP peers:
– two LSRs which use LDP to exchange label/stream
mapping
– information exchange known as “LDP Session”

15
Copyright © 2013 Tech Mahindra. All rights reserved. 15
 LDP Message Exchange
 Discovery messages - used to announce and maintain
the presence of an LSR
 Session messages - used to establish, maintain and
terminate sessions between LDP peers
 Advertisement messages - used to create, change,
and delete label mappings
 Notification messages - used to provide advisory
information and to signal error information

16
Copyright © 2013 Tech Mahindra. All rights reserved. 16
LDP Message Format

0 1 2 3

01234567890123456789012345678901

U Message Type Message Length

Message ID

Mandatory Parameters

Optional Parameters

17
Copyright © 2013 Tech Mahindra. All rights reserved. 17
 LDP Protocol Data Units (PDUs)
 LDP message exchanges are accomplished by
sending LDP PDUs
 Each LDP PDU is an LDP header followed by LDP
message
 The LDP header is:
0 1 2 3

01234567890123456789012345678901

Version PDU Length

LDP Identifier

18
Copyright © 2013 Tech Mahindra. All rights reserved. 18
Forwarding Equivalence Class (FEC)
 Introduced in MPLS standards to denote packet
forwarding classes
 Comprises traffic
 Why FEC?
– To precisely specify wto a particular destination
– to destination with distinct service requirements
– hich IP packets are mapped to each LSP
– Done by providing a FEC specification for each
LSP

19
Copyright © 2013 Tech Mahindra. All rights reserved. 19
LSP - FEC Mapping

 FEC specified as a set of two elements (currently)

1. IP Address Prefix - any length from 0 - 32
2. Host Address - 32 bit IP address
 A given packet matches a particular LSP if and only if IP
Address Prefix FEC element matches packet’s IP destination
address

20
Copyright © 2013 Tech Mahindra. All rights reserved. 20
 Rules for Mapping packet to a LSP
 If exactly one LSP’s Host Address FEC element ~ packet’s
IP destination address, packet is mapped to that LSP
 If there are multiple LSPs satisfying the above condition,
then the packet is mapped to one of those LSPs†
 If a packet matches exactly one LSP, packet is mapped to
that LSP
 If packet matches multiple LSPs, mapped to one with the
longest prefix match

21
Copyright © 2013 Tech Mahindra. All rights reserved. 21
 Label Spaces

 Useful for assignment and distribution of labels

 Two types of label spaces
 Per interface label space: Interface-specific labels used for interfaces
that use interface resources for labels
 Per platform label space: Platform-wide incoming labels used for
interfaces that can share the same label space

22
Copyright © 2013 Tech Mahindra. All rights reserved. 22
LDP Identifiers
 A six octet quantity
 used to identify specific label space within an LSR
 First four octets encode LSR’s IP address
 Last two octets identify specific label space
 Representation <IP address> : <label space id>
– e.g., 171.32.27.28:0, 192.0.3.5:2
 Last two octets for platform-wide label spaces are
always both zero

23
Copyright © 2013 Tech Mahindra. All rights reserved. 23
LDP Discovery
 A mechanism that enables an LSR to discover potential
LDP peers
 Avoids unnecessary explicit configuration of LSR label
switching peers
 Two variants of the discovery mechanism
– basic discovery mechanism: used to discover LSR
neighbors that are directly connected at the link level
– extended discovery mechanism: used to locate LSRs
that are not directly connected at the link level

24
Copyright © 2013 Tech Mahindra. All rights reserved. 24
LDP Discovery contd..
 Basic discovery mechanism
– To engage - send LDP Hellos periodically
– LDP Hellos sent as UDP packets for all routers on that
subnet
 Extended discovery mechanism
– To engage - send LDP targeted Hellos periodically
– Targeted Hellos are sent to a specific address
– Targeted LSR decides whether to respond or to ignore the
targeted Hello
 LDP Link Hello sent by an LSR
– carries the LDP identifier for the label space the LSR
intends to use for the interface

25
Copyright © 2013 Tech Mahindra. All rights reserved. 25
 Session establishment
 Exchange of LDP discovery Hellos triggers session establishment
 Two step process
 Transport connection establishment
– If LSR1 does not already have a LDP session for the exchange of label spaces
LSR1:a and LSR2:b, it attempts to open a TCP connection with LSR2
– LSR1 determines the transport addresses at it’s end (A1) and LSR2’s end (A2) of
the TCP connection
– If A1>A2, LSR1 plays the active role; otherwise it is passive
 Session initialization
– Negotiate session parameters by exchanging LDP initialization messages

26
Copyright © 2013 Tech Mahindra. All rights reserved. 26
 Diagram
NON EXISTENT
Session connection Rx Any LDP msg except
Rx Any other established Init msg or Timeout
msg or (Passive
Timeout Role) INITIALIZED (Active
Tx NAK msg Rx Role)
Acceptable
Tx Init msg
Init msg/
Tx Init msg &
KeepAlive
msg
OPENREC OPENSENT Rx Any other
Rx KeepAlive msg or
Timeout
msg Rx Acceptable Init msg Tx NAK msg
Tx KeepAlive msg
OPERATIONAL Rx Shutdown
All other LDP msgs msg or Timeout
Tx Shutdown msg
Rx - Receive
Tx - Transmit
Copyright © 2013 Tech Mahindra. All rights reserved. 27
Session Initialization State Transition
Table
STATE EVENT NEW STATE

NON Session TCP connection established INITIALIZED

EXISTENT
INITIALIZED Transmit initialization message OPENSENT
(Active Role)

Receive acceptable initialization OPENREC

message (Passive role)
Action: Transmit initialization
message and Keep alive message

Receive Any other LDP msg NON

Action: Transmit error notification EXISTENT
msg (NAK) and close transport
connection
28
Copyright © 2013 Tech Mahindra. All rights reserved. 28
Session Initialization State Transition Table (cont

STATE EVENT NEW STATE

OPENREC Receive KeepAlive msg OPERATIONAL
Receive Any other LDP msg
Action: Transmit Error NON EXISTENT
Notification msg (NAK) and close
transport connection
OPENSENT Receive acceptable Init msg OPENREC
Action: Transmit KeepAlive msg
NON EXISTENT
Receive Any other LDP msg
Action: Transmit Error msg
(NAK) and close transport
connection

OPERATIONAL Receive Shutdown msg NON EXISTENT

Action: Transmit Shutdown msg
and close transport connection

All other LDP messages OPERATIONAL

Copyright © 2013 Tech Mahindra. All rights reserved. 29
 Label Distribution and Management

 Two label distribution techniques

 Downstream on demand label distribution:
An LSR can distribute a FEC label binding in response to an explicit request
 Downstream Unsolicited label distribution:
Allows an LSR to distribute label bindings to LSRs that have not explicitly
requested them
 Both can be used in the same network at the same time; however, each LSR must
be aware of the distribution method used by its peer

30
Copyright © 2013 Tech Mahindra. All rights reserved. 30
 Label Distribution Control Mode

 Independent Label Distribution Control

 Each LSR may advertise label mappings to its neighbors at any time
 In independent Downstream on Demand mode - LSR answers without
waiting for a label mapping from next hop
 In independent Downstream Unsolicited mode - LSR advertises label
mapping for a FEC whenever it is prepared
 Consequence: upstream label can be advertised before a downstream
label is received

31
Copyright © 2013 Tech Mahindra. All rights reserved. 31
Label Distribution Control Mode
contd..
 Ordered Label Distribution Control
 Initiates transmission of label mapping for a FEC only if it has next FEC next
hop or is the egress
 If not, the LSR waits till it gets a label from downstream LSR
 LSR acts as an egress for a particular FEC, if
– next hop router for FEC is outside of label switching network
– FEC elements are reachable by crossing a domain boundary

32
Copyright © 2013 Tech Mahindra. All rights reserved. 32
 Label Retention Mode

 Conservative Label Retention Mode

 Advertised label mappings are retained only if they are used for
forwarding packets
 Downstream on Demand Mode typically used with Conservative Label
Retention Mode
 Advantage: only labels required are maintained
 Disadvantage: a change in routing causes delay
 Liberal Retention Mode
 All label mappings are retained regardless of whether LSR is next hop or
not
 reaction to routing changes will be quick

33
Copyright © 2013 Tech Mahindra. All rights reserved. 33
Label Information Base

 LSR maintains learned labels in Label Information Base (LIB)

 Each entry of LIB associates an FEC with an (LDP Identifier, label) pair
 When next hop changes for a FEC, LSR will retrieve the label for the new next
hop from the LIB

34
Copyright © 2013 Tech Mahindra. All rights reserved. 34
 Hierarchical Operation in MPLS

Example:
•External Routers A,B,C,D,E,F - Talk BGP
•Internal Routers 1,2,3,4,5,6 - Talk OSPF

C D
Domain #2
1 6
2 3 4 5
A B E F
Domain #1 Domain #3

Note: Internal routers in domains 1 and 3 not shown 35

Copyright © 2013 Tech Mahindra. All rights reserved. 35
 Hierarchical Operation contd..

 When IP packet traverses domain #2, it will contain two labels, encoded as a “label
stack”
 Higher level label used between routers C and D, which is encapsulated inside a
lower level label used within Domain #2
 Operation at C
 C needs to swap BGP label to put label that D expects
 C also needs to add an OSPF label that 1 expects
 C therefore pushes down the BGP label and adds a lower level label

36
Copyright © 2013 Tech Mahindra. All rights reserved. 36
 Label Stack

 Multiple labels are carried in data packets

 e.g. data packet carried across Domain #2
 Concept of stacking
 provides a mechanism to segregate streams within a switched path
 one useful application of this technique is in Virtual Private Networks
 Advantage of Hierarchical MPLS is that the internal routers need not
know about higher level (BGP) routing

37
Copyright © 2013 Tech Mahindra. All rights reserved. 37
 Multipath

 Many IP routing protocols support the notion of equal-cost multipath routes

 Few possible approaches for handling multipath within MPLS
 First approach:
 separate switched path from each ingress node to the merge point
 preserves switching performance, but at the cost of proliferating the number
of switched paths

38
Copyright © 2013 Tech Mahindra. All rights reserved. 38
 Multipath contd..
 Second approach
 Only one switched path from one ingress node to a destination
 Conserves switched paths but cannot balance loads across downstream links as well
as other approaches
 LSP may be different from the normal L3 path
 Third approach:
 Allows single stream to be split into multiple streams, by using L3 forwarding
 e.g. might use a hash function on source and destination IP addresses
 Conserves paths at the cost of switching performance

Copyright © 2013 Tech Mahindra. All rights reserved. 39

 Explicit Routing in MPLS

 Two options for route selection:

 Hop by hop routing
 Explicit routing
 Explicit Routing (aka Source Routing) is a very powerful technique
 With pure datagram routing overhead of carrying complete explicit route is
prohibitive
 MPLS allows explicit route to be carried only at the time the LSP is setup,
and not with each packet
 MPLS makes explicit routing practical

40
Copyright © 2013 Tech Mahindra. All rights reserved. 40
 Explicit Routing in MPLS contd..

 In an explicitly routed LSP

 the LSP next hop is not chosen by the local node
 selected by a single node, usually the ingress
 The sequence of LSRs may be chosen by
 configuration (e.g., by an operator or by a centralized server)
 an algorithm (e.g., the ingress node may make use of topological
information learned from a link state routing protocol)

41
Copyright © 2013 Tech Mahindra. All rights reserved. 41
Loops and Loop Handling

 Routing protocols used in conjunction with MPLS are based on distributed

computation which may contain loops
 Loops handling - 3 categories
 Loop Survival
 Loop Detection
 Loop Prevention

42
Copyright © 2013 Tech Mahindra. All rights reserved. 42
 Loop Survival

 Minimizes the impact of loops by limiting the amount of resources

consumed by the loop
 Method
 based on use of TTL field which is decrement at each hop
 Use of dynamic routing protocol converging rapidly to non-looping paths
 Use of fair queuing

43
Copyright © 2013 Tech Mahindra. All rights reserved. 43
Loop Detection

 Loops may be setup but they are subsequently detected

 The detected loop is then broken by dropping label relationship
 Broken loops now necessitates packets to be forwarded using L3 forwarding

44
Copyright © 2013 Tech Mahindra. All rights reserved. 44
Loop Detection (cont.)

 Method is based on transmitting a Loop Detection Control Packet (LDCP)

whenever a route changes
 LDCP is forwarded towards the destination until
 last MPLS node along the path is reached
 TTL of the LDCP expires
 it returns to the node which originated it

45
Copyright © 2013 Tech Mahindra. All rights reserved. 45
Loop Prevention

 Ensures that loops are never set up

 labels are not used until it is sure to be loop free
 Methods
 labels are propagated starting at the egress switch
 use source routing to set up label bindings from the egress switch to each
ingress switch

46
Copyright © 2013 Tech Mahindra. All rights reserved. 46
Leaf Leaf
Leaf

Detects loop
immediately

Link removed Change in Link

from tree
LSR

Ingress Node

Egress Node 47
Copyright © 2013 Tech Mahindra. All rights reserved. 47
 Traffic Engineering and Performance
Objectives
 Traffic Engineering (TE) is concerned with performance optimization of
operational networks
 The key performance objectives
 traffic oriented - aspects that enhance the QoS of traffic streams e.g
minimization of packet loss
 resource oriented - aspects that pertain to the optimization of resource
utilization e.g efficient management of bandwidth

48
Copyright © 2013 Tech Mahindra. All rights reserved. 48
 Performance Objectives (cont.)

 Minimizing congestion is a major traffic and resource oriented

performance objective
 Congestion manifest under two scenarios
 network resources are insufficient or inadequate
– can be solved by capacity expansion or classical congestion control
techniques
 traffic streams are inefficiently mapped onto available resources
– can be reduced by adopting load balancing policies

49
Copyright © 2013 Tech Mahindra. All rights reserved. 49
 Traffic and Resource Control

 The traffic engineer acts as the controller in an adaptive feedback control

system which includes
 a set of interconnected network elements
 a network performance monitoring system &
 network configuration management tools
 The traffic engineer formulates control policies, observes the state of the
network, characterizes the traffic and applies the control actions in
accordance to the control policy

50
Copyright © 2013 Tech Mahindra. All rights reserved. 50
 MPLS and Traffic Engineering

 Main components used

 Traffic Trunk - aggregation of traffic flows of the same class which are
placed inside a Label Switched Path
 Induced MPLS Graph
– analogous to a virtual topology in an overlay model
– logically mapped onto the physical network through the selections o
LSPs for traffic trunk
– comprises a set of LSRs which act as nodes of the graph and a set of
LSPs which provide logical point to point connectivity between LSRs
and thus act as edges of the graph

51
Copyright © 2013 Tech Mahindra. All rights reserved. 51
 Augmented Capabilities

 Set of attributes associated with traffic trunks which collectively specify

their behavioral characteristics
 Set of attributes associated with resources which constrain the
placement of traffic trunks through them
 A “constraint based routing” framework which is used to select paths for
traffic trunks subject to constraints imposed

52
Copyright © 2013 Tech Mahindra. All rights reserved. 52
 Basic operation on traffic trunks

 Establish - create an instance of a traffic trunk

 Activate - cause to start passing traffic
 Deactivate - stop passing traffic
 Modify Attributes
 Reroute - administratively or by underlying protocols
 Destroy - reclaim all resources such as label space and bandwidth

53
Copyright © 2013 Tech Mahindra. All rights reserved. 53
 Basic attributes of traffic trunk

 Traffic parameter attribute - capture the characteristics of the traffic streams

 Generic Path selection and maintenance attributes - defines rules for selecting
route taken by traffic trunk and rules of maintaining the paths
 Priority attribute
 Preemption attribute
 Resilience attribute
 Policing attribute

54
Copyright © 2013 Tech Mahindra. All rights reserved. 54
 Resource Attributes

 Part of the topology state parameters used to constrain the routing of traffic trunks
through specific resources
 Main components
 Maximum Allocation Multiplier (MAM) - administratively configured to determine the
proportion of resource available for allocation
 Resource Class Attribute - administratively assigned parameters which express some
notion of “Class” for resources

55
Copyright © 2013 Tech Mahindra. All rights reserved. 55
 Constraint Based Routing

 Enables a demand driven, resource reservation aware, routing paradigm to co-

exist with current topology driven protocols
 uses the following inputs
 traffic trunk attributes
 resource attributes
 other topology state information
 Basic features
 prune the resources that do not meet the requirements of the traffic trunk attribute
 run a shortest path algorithm on the residual graph

56
Copyright © 2013 Tech Mahindra. All rights reserved. 56
Constraint Based Routing (cont.)

 Strict & Loose Explicit Routes

 Constraint Based LSP (CRLSP) is calculated at one point at the edge of the
network based on certain criteria
 special char. such as assigning certain bandwidth can be supported
 The route is encoded as a series of Explicit routed hops contained in a CR
based route TLV

57
Copyright © 2013 Tech Mahindra. All rights reserved. 57
 Constraint Based Routing (cont.)

 Traffic Characteristics
 Described in the Traffic Parameter TLV in terms of peak rate, committed
rate and service granularity
 Preemption
 Setup and Holding priorities are used to rank new and existing paths
respectively to determine if new paths can preempt existing paths
 Allocation of these priorities is a network policy

58
Copyright © 2013 Tech Mahindra. All rights reserved. 58
 Constraint Based Routing (cont.)

 Route Pinning
 applicable to segments of an LSP that are loosely routed i.e the next hop
is an abstract node
 used if the LSP need not be changed
 Resource Class
 While setup , indication must be given as to which class the CRLSP can
draw resources from

59
Copyright © 2013 Tech Mahindra. All rights reserved. 59
 Implementation Consideration

Management Interface

Constraint Based Conventional

MPLS
Routing Process IGP Process

Resource Attribute Link State

Availability Database Database
60
Copyright © 2013 Tech Mahindra. All rights reserved. 60
 Quality of Service using CRLSP

 Delay Sensitive Service

 the network commits to deliver with high probability, user datagrams at a rate of PDR
with minimum delay and delay requirements
 Datagrams in excess of PDR will be discarded
 Throughput Sensitive Service
 the network commits to deliver at a rate of at least CDR
 Datagrams with higher CDR have lower probability of being delivered
 Best Effort Service
 No expected service is guaranteed

61
Copyright © 2013 Tech Mahindra. All rights reserved. 61
 Tag Switching

Terminologies Analogies in Label

Switching
Tags Labels

Tag Switching Router (TSR) Label Switching Router

Tag Edge Router (TER) Edge Label Switching

Router
Tag Forwarding Information Label Switching Forwarding
Base (TFIB) Table
Tag Distribution Protocol Label Distribution Protocol
(TDP)
62
Copyright © 2013 Tech Mahindra. All rights reserved. 62
Destination Based Routing

 A TSR participates in unicast routing protocols to construct its mapping

between FECs and next hops
 This mapping is used by the Tag Switching Control component for
constructing the TFIB which is used for actual packet forwarding

63
Copyright © 2013 Tech Mahindra. All rights reserved. 63
 Destination Based forwarding model of
Tag Switching

A B

if0 if1 if2

if1 E
if2 if0
if0

if1 if2 192.16/16

if0 if2 if1 if0 TSR

C D

64
Copyright © 2013 Tech Mahindra. All rights reserved. 64
 Information for constructing TFIB

 A local binding between the FEC and a tag

 takes a tag from the pool of free tags and uses it as an index in the TFIB to
set the incoming tag entry
 A mapping between the FEC and the next hop for that FEC (provided by
the routing protocol(s) running on the TSR)
 A remote binding between the FEC and a tag that is received from the next
hop

65
Copyright © 2013 Tech Mahindra. All rights reserved. 65
 Initial TFIB Entries

Incoming Outgoing Next Outgoing

tag tag hop Interface
On TSR A 100 ? TSR B If1
On TSR B 6 ? TSR E If1
On TSR C 17 ? TSR D If2
On TSR D 5 ? TSR E If0
On TSR E 6 ? TSR E If0

66
Copyright © 2013 Tech Mahindra. All rights reserved. 66
TFIB Entries after Tag Distribution

Incoming Outgoing Next Outgoing

tag tag hop Interface
On TSR A 100 6 TSR B If1
On TSR B 6 6 TSR E If1
On TSR C 17 5 TSR D If2
On TSR D 5 6 TSR E If0
On TSR E 6 ? TSR E If0

67
Copyright © 2013 Tech Mahindra. All rights reserved. 67
 Behavior during routing change

A B

if0 if1 if2

if1 E
if2 if0
if0

if1 if2

Link Down

if0 if2 if1 if0 TSR

C D

68
Copyright © 2013 Tech Mahindra. All rights reserved. 68
 Updated TFIB

Incoming Outgoing Next Outgoing

tag tag hop Interface
On TSR A 100 6 TSR B If1
On TSR B 6 6 TSR E If1
On TSR C 17 5 TSR D If2
On TSR D 5 6 TSR B If0
On TSR E 6 ? TSR E If0

69
Copyright © 2013 Tech Mahindra. All rights reserved. 69
 Hierarchy of Routing Knowledge

 All TSRs within a routing domain participate in a common intra-domain

routing protocol and construct TFIB corresponding to destinations within
the domain
 All border TSRs or TERs within a domain and directly connected TERs
from other domains also exchange Tag binding information via inter-domain
routing protocol

70
Copyright © 2013 Tech Mahindra. All rights reserved. 70
 Hierarchy of Routing Knowledge (cont.)

 To support forwarding in the presence of hierarchy of routing knowledge, Tag

switching allows a packet to carry several tags organized as a tag stack
 At the ingress a tag is pushed onto the tag stack, and at the egress a tag is
popped off a the stack

71
Copyright © 2013 Tech Mahindra. All rights reserved. 71
 Hierarchy of Routing knowledge model

Routing Routing
domain Routing domain A domain
B C

V T X Y W Z

TSR

72
Copyright © 2013 Tech Mahindra. All rights reserved. 72
 TFIB Entries in Routing Domain A

Incoming Outgoing Next

tag tag hop
On TSR A N/A 10 TSR X
On TSR B 10 12 TSR Y
On TSR C 12 17 TSR W
On TSR D 17 N/A TSR W

73
Copyright © 2013 Tech Mahindra. All rights reserved. 73
 Label Stack During Hierarchical Routing

TSR Z distributes label 2 to TSR W and TSR W gives

label 5 to TSR T for the purpose of inter-domain routin

Top of
Stack
10 Top of
Stack

2 2
Stack after processing in Stack after processing in
TSR T TSR W
74
Copyright © 2013 Tech Mahindra. All rights reserved. 74
Multicast in Tag Switching

 Selects the distribution tree based only on

 tag carried in a packet
 interface on which the packet arrives
 TSR maintains its TFIB on a per interface basis
 TSRs connected to a common sub-network agree among themselves on a
common tag associated with a particular multicast tree

75
Copyright © 2013 Tech Mahindra. All rights reserved. 75
 Multicast in Tag Switching (cont.)

 Procedures are used to partition the set of tags for use with multicast into
disjoint subsets and care is taken to avoid overlapping with the help of
HELLO packets
 TSR connected to a common sub-network and those which are a part of
the same distribution tree elect one TSR that will create the tag bindings
and distribute them and any TSR can join the group using the JOIN
command

76
Copyright © 2013 Tech Mahindra. All rights reserved. 76
 Multicast model in Tag Switching

A B
TSR

if0
D
if0 if1

if2

if0 if0

E 77 F
Copyright © 2013 Tech Mahindra. All rights reserved. 77
 RSVP with Tag Switching

 RSVP is supported by the help of a RSVP object - the tag Object

 The tag object binding information for an RSVP flow is carried in the RSVP
“RESV” message
 The RESV message carries the tag object containing the tag given by a
TSR and also information about the local resources to be used
 The reservation state is refreshed once the flow is set up using the RESV
message

78
Copyright © 2013 Tech Mahindra. All rights reserved. 78
Explicit Routes

 Tag switching supports explicit routes with the help of a RSVP object - the
Explicit Route Object
 The object is carried in the RSVP “PATH” message
 The tag information is carried in the Tag Object by the RSVP “RESV”

79
Copyright © 2013 Tech Mahindra. All rights reserved. 79
 IP Switching

 Introduced by Ipsilon
 Already been tested in the field
 Significant Innovation: Defined a switch management protocol (GSMP)
along with label binding protocol called Ipsilon Flow Management Protocol
(IFMP)
 General Switch Management Protocol (GSMP) - allows an ATM switch to
be controlled by an “IP switch controller”

80
Copyright © 2013 Tech Mahindra. All rights reserved. 80
 IP Switching Overview

 IP over ATM models are complex and inefficient - involve running two
control planes
 ATM Forum signaling and routing
 IP routing and address resolution on top
 In contrast IP Switching uses
 IP component plus label binding protocol
 completely removes ATM control plane
 Goal: To integrate ATM switches and IP routing in a simple and efficient
way

81
Copyright © 2013 Tech Mahindra. All rights reserved. 81
 Removing ATM Control Plane

IP
ATM MARS NHRP
ARP
PNNI
IP IFMP
Q.2931

ATM hardware ATM hardware

(a) (b)

(a) IP over Standard ATM

(b) IP Switching

82
Copyright © 2013 Tech Mahindra. All rights reserved. 82
 IP Switching Architecture

 Switch controller
 control processor of the system
 uses GSMP to communicate with ATM switch itself
 runs IP routing and forwarding code
 Default VC
 defined to get control traffic before IP Switching is performed
 uses well known VCI/VPI value
 also used for data that doesn’t yet have a label

Copyright © 2013 Tech Mahindra. All rights reserved. 83

 IP Switch Architecture

Switch controller
Flow Classification and control
To
To Routing downstream

GSMP
IFMP
upstream and switch
switch forwardin
g

GSMP
Default Default
VC VC
Data Data
VC VC
Switch
84
Copyright © 2013 Tech Mahindra. All rights reserved. 84
 IP Switching Basics

 IP Switching relies on IP protocols

 to establish routing information
 to determine next hop
 Flow classification and control module selects flows from incoming traffic
 IP flow refers to a sequence of datagrams
 from one source to one destination, identified by the ordered pair <source address,
destination address>
 can also refer to a flow at finer granularity, e.g., different applications between same
pair of machines, identified by < source address, source port, destination address,
destination port>

Copyright © 2013 Tech Mahindra. All rights reserved. 85

 Flow Redirection

 Redirection: Process of binding labels to flows and establishing label

switched paths
 Example:
 data is flowing from A via B to C on default VC
 B sends a redirect to A specifying flow y and the label (VPI/VCI) on which it
expects to receive
 If C issues a redirect to B for flow y, B forwards y on the VPI/VCI specified by
C
 Since same flow y enters B on one VC and leaves on another, B uses GSMP
to inform its switching element to set up the appropriate switching path

Copyright © 2013 Tech Mahindra. All rights reserved. 86

 Flow Redirection
Redirect:
Flow y VPI/VCI 3/57
Switch
A B C Controller
Default Default
VC VC Switch
3/57 Element

Switch B issues a REDIRECT message to switch A

Redirect: Redirect:
Flow y VPI/VCI 3/57 Flow y VPI/VCI 2/22
Switch
A B C Controller
Default Default
VC VC Switch
Element
3/57 2/22
Switch B and C redirect the same flow, allowing it to be switched at B
Copyright © 2013 Tech Mahindra. All rights reserved. 87
 Ipsilon Flow Management Protocol
(IFMP)

 Designed to communicate flow to label binding information

 IFMP is a soft state protocol
 IFMP’s Adjacency Protocol:
 Used to communicate and discover information about neighbors
 Adjacency message sent as limited broadcast
 IFMP’s Redirection Protocol
 used to send appropriate messages for flow-label bindings

88
Copyright © 2013 Tech Mahindra. All rights reserved. 88
 IFMP’s Redirection Protocol

 Different message types defined:

 REDIRECT: used to bind label to a flow
 RECLAIM: enables label to be unbound for subsequent re-use
 RECLAIM ACK: Acknowledgement for RECLAIM message
 ERROR: Used to deal with various error conditions
 Common header format

89
Copyright © 2013 Tech Mahindra. All rights reserved. 89
FMP Redirect Protocol Message Format
Version Op code Checksum

Sender Instance

Peer Instance

Sequence Number

Message body: variable length

IFMP REDIRECT message body

Flow type Flow ID length Lifetime

Label

Flow identifier

Copyright © 2013 Tech Mahindra. All rights reserved. 90

 Encapsulation of Redirected Flows

LLC SNAP IP header Data AAL5

trailer
Encapsulation of IP packet on the default VC

IFMP flow Data AAL5

type header trailer
Encapsulation of IP packet on the redirected VCs

91
Copyright © 2013 Tech Mahindra. All rights reserved. 91
 General Switch Management Protocol
(GSMP)
 GSMP is a master/slave protocol
 ATM switch is the slave
 Master could be any general purpose computer
 The protocol allows the master to
 Establish and release VC connections across the switch
 Perform port management (Up, Down, Reset, Loopback)
 Request Data (configuration information, statistics)
 Allows slave to inform master if something interesting, such as link failure, happens
on the switch

92
Copyright © 2013 Tech Mahindra. All rights reserved. 92
 GSMP contd..

 GSMP packets are LLC/SNAP encapsulated and sent over ATM link using
AAL5
 GSMP Adjacency Protocol
 used to gain information about the system at the other end of the link and
 to monitor link status
 GSMP Connection Management Protocol
 used to ensure consistency between the GSMP master and slave
 also specifies the QoS using a priority field

93
Copyright © 2013 Tech Mahindra. All rights reserved. 93
 Implementations & Contributions

 IP Switching products
 available since 1996
 Ipsilon product family uses Intel Pentium-based PC as the switch controller
 Also offers a number of ATM switches that are controlled by the switch controller
 IP Switching made the following significant contributions to label switching effort:
 first to deliver real products and caused activity that resulted in the development of
Tag Switching and ultimately the formation of MPLS working group
 contributed GSMP

94
Copyright © 2013 Tech Mahindra. All rights reserved. 94