第二章下一代网络与 mpls chapter 2 next generation network and mpls

1

现代通信新技术导论

第二章下一代网络与 MPLSChapter 2 Next Generation Network and MPLS

电控学院电子工程学科部司鹏搏综合楼 825 室[email protected]

现代通信新技术导论第二章下一代网络与 MPLS

2

Main Contents• Next Generation Network

– Background of Next Generation Network (NGN)

– Definition of NGN

– Characteristics of NGN

– Architectures of NGN• 3GPP

• ETSI

• With Softswitch

• MPLS


3


• MPLS– Need for MPLS

– MPLS Architecture• Label Switching

• MPLS Basics

– MPLS Labels• Label Forwarding

• Label Distribution and Management

• MPLS Forwarding Table

• LSP Loop Detection

• LDP


4






• ETSI

• With Softswitch

• MPLS


5

Background of NGN• Limited Capacity of Traditional Internet

– Simple services are not sufficient• Voice, text message, e-mail

– Non-open system

– Different networks are independent

– Old technologies are not ready for evolution

• Why Next Generation Network?– The fast growing of data service requirement

– The slow growing/decreasing of voice service

– The increasingly cut-throat competition

– The rapid development of new technologies


6

Non-Interactive Multimedia

Image

SMS

MMS

Presence

Activephonebook

Push-To-Talk

Text

Voice

Voice

P2P Calls

Video

Person-to-Person

dominates traffic growth

Movies

PhotosInternet

Text/Pictures

SMS/MMS

HTTP

Streaming

Download

VideoMusic

Ring tonePerson-to-Content

known usability patterns

Interactive Multimedia

Multimedia Content

Social Networking


7


8

Societal and Business trends• Internet is becoming a major enabler of communications

• Consumers are embracing computing, mobile and digital technology in their everyday life

• Evolution of Business models require increased levels of personal mobility

Societal and Business trends• Internet is becoming a major enabler of communications

• Consumers are embracing computing, mobile and digital technology in their everyday life

• Evolution of Business models require increased levels of personal mobility

Access Technology Enhancements• HSPA (High Speed Packet Access) – evolved WCDMA

• OFDMA (Orthogonal Frequency Division Multiple Access) – 3GPP LTE, WiMAX, MBWA, ADSL/VDSL, DVB-T/H etc.

• Spatial Processing – multi-antennas Base Stations supporting advanced spatial processing

Access Technology Enhancements• HSPA (High Speed Packet Access) – evolved WCDMA

• OFDMA (Orthogonal Frequency Division Multiple Access) – 3GPP LTE, WiMAX, MBWA, ADSL/VDSL, DVB-T/H etc.

• Spatial Processing – multi-antennas Base Stations supporting advanced spatial processing

Convergence• Converged devices (Mobile, WLAN, Internet etc.) Connectivity

• Converged services Ease of use

• Converged networks Reliability, Security, Reduced OPEX/CAPEX

• Converged business models Increased margins, Avoidance of twin pitfalls risk

Convergence• Converged devices (Mobile, WLAN, Internet etc.) Connectivity

• Converged services Ease of use

• Converged networks Reliability, Security, Reduced OPEX/CAPEX

• Converged business models Increased margins, Avoidance of twin pitfalls risk


9






• ETSI

• With Softswitch

• MPLS


10

Definition of NGN• By ITU-T

– In Feb. 3 - 12, 2004, ITU-T SG13

– A Next Generation Network is a packet-based network able to provide telecommunication services, able to make use of multiple broadband, QoS-enabled transport technologies and in which service-related functions are independent from underlying transport-related technologies.

– It offers unrestricted access by users to different service providers.

– It supports generalized mobility which will allow consistent and ubiquitous provision of services to users.


11

Definition of NGN

Packet Oriented Network

Integration of Exist-ing Infrastructure

Application Fo-cused - Access

Independent

Seperation into Dif-ferent Layers Using

Open Interfaces

Openess and Flex-ibility Regarding New Services

Support Broad Va-riety of Services

NGN CONCEPT


12






• ETSI

• With Softswitch

• MPLS


13

Characteristics of NGN

• Packet-based transfer

• Separation of control functions

• Decoupling of service provision

• Support for a wide range of services

• Broadband capabilities with end-to-end QoS

• Interworking via open interfaces

• Generalized mobility


14

Characteristics of NGN

• Unrestricted access

• A variety of identification schemes

• Unified service characteristics

• Converged services between Fixed/Mobile

• Independence of service-related functions

• Support of multiple RATs

• Compliant with all regulatory requirements


15






• ETSI

• With Softswitch

• MPLS


16

Architecture Evolution

Traditional Architecture


17



18



19


Phone

PSTN

Internet

IP-Net

M-Ph

Mobile

Pol. AReg. a

Current PolicyRegulationEnvironment (Vertical)

Pol. BReg. b

Pol. CReg. c

Resource-based Pol./Reg.

Video

NGN

VoiceInternet

IP (Future Packet ?) Platform

xDSL/Optic based Fixed-Mobile

MMPol.Reg.

Pol. XReg. x

New PolicyRegulationEnvironment (Horizontal)

Service-based Pol./Reg.


20



21



22

The Players of NGN

ITU-T NGN FG

ATIS NGN FG

ETSI TISPAN1

3GPP


23






• ETSI

• With Softswitch

• MPLS


24

3GPP Architecture of NGN with IMS


25

• IMS is an open IP-based architecture using the Client-Server Network Computing model– 3GPP originally specified IMS to enable real-time multimedia services over the IP

bearer, in GSM and WCDMA networks.

– 3GPP2 specified the MMD architecture for CDMA2000 networks based on IMS. 3GPP2 requirements are part of Common IMS in IMS release 8.

– The xDSL access, specified by TISPAN, is integrated into IMS.

– The cable access specified by CableLabs in PacketCable 2.0 is part of IMS release 8.

– Interworking with WLAN was specified in IMS release 6, while the mobility with WiMAX has been addressed in EPC specifications.

• If IMS is not used:– Multimedia communication at best effort

– Service roaming can be difficult to implement

– Provisioning and charging are service specific

– Compliance with LI requirements can be an issue

3GPP Architecture of NGN with IMS


26






• ETSI

• With Softswitch

• MPLS


27

ETSI TISPAN NGN Architecture


28

NGN Architecture Principles• A Layered Approach Including

– A transport layer, including functional entities that do transport routing

– A service layer, including functional entities that provide services

• A Sub-System Oriented Approach Enabling– The addition of new subsystems over the time to cover new

demands and service classes

– To import (and adapt) subsystems from other standardization bodies

– Flexibility to adjust a subsystem architecture with no or limited impact on other subsystems


29

NGN Architecture Principles• IP Connectivity is Provided Using Two Sub-

Systems:– Network Attachment Subsystem (NASS)

– Resource and Admission Control Subsystem (RACS)

• First Service-Oriented Sub-Systems Include– the 3GPP IMS, a PSTN/ISDN Simulation Sub-system

– a PSTN/ISDN Emulation Subsystem (PES)

• Future Service-Oriented Sub-Systems may Include– A streaming subsystem

– A TV Broadcasting subsystem


30


31

Common Components• Common components are functions that are used by more

than a subsystem– USPF: User Profile Server Function

– Service-level user identification, numbering, and addressing information

– Service-level user security information

– Service-level user location information

– Service-level user profile storage

– The IMS part of the USPF is similar to the 3GPP HSS without the HLR/AUC


32

Common Components• ASF: Application Server Function

– Offers services

– Two types of ASFs:• Type 1: may interact with RACS for resource control purposes

• Type 2: relay on the control subsystem.– Type 2 is equivalent to the Application Server defined by 3GPP IMS

• SLF: Subscription Locator Function– Helps locating the USPF of a given user

– Like the SLF defined by 3GPP IMS


33

Common Components• Charging Functions

– Data collection functions and mediation functions to the billing systems

– Both on-line and off-line charging

• IWF: Interworking Function– Between different SIP profiles

– Between SIP and H.323

• IBCF: Interconnection Border Control Function– Controls an operator’s boundary

– Interacts with RACS

– Inserts IWF when appropriate

– Screening of signaling based on source/destination addresses


34


35

Transfer Functions


36

Transfer Functions• BGF: Border Gateway Function

– Interface between two IP transport domains

– Open/close gates, packet marking, resource allocation, bandwidth reservation, NAPT, NAT traversal, incoming traffic policing, usage metering, IPv4/IPv6 interworking, topology hiding

– Three types of BGF:• A-BGF (Access-BGF): located in between access and core network, at the

access network side

• C-BGF (Core-BGF): located in between access and core network, at the core network side

• I-BGF (Interconnection-BGF): located in between two core networks


37

Transfer Functions• L2TF: Layer 2 Termination Function

– Terminates Layer 2, e.g., PPP or DSLAM

• ARF: Access Relay Function– Relay between the CPE and the NASS

– Located in the access network

– Resource reservation and admission control

• SGF: Signaling Gateway Function– Conversion of SS7 protocols to IP control protocols (e.g., SIP)

– SS7 screening of MTP and SCCP parameters


38

Transfer Functions• MGF: Media Gateway Function

– Media mapping and transcoding between IP and CS networks

– Three types of MGF:• R-MGF Residential MGF: located in the customer premises

• A-MGF Access MGF: located in the access or core network

• T-MGF Trunking: MGF located in the boundary of the core network and PSTN/ISDN network

• MRFP: Media Resource Function Processor– Multimedia conferences, media sourcing, IVR capabilities,

media content analysis


39


40

RACS Functionality• Support for Two Types of QoS:

– Guaranteed QoS: resources are reserved

– Support for Relative QoS: diffserv marking

• Service Based Local Policy Control: authorization of QoS requests and definition of the polices to be enforced by the bearer service network elements.– Resource reservation

– Support for two mechanisms• Application Function (AF)-initiated

• CPE-initiated (Authorization token a la 3GPP).

– QoS support over multiple access networks (e.g. ADSL and GPRS) and CPE types.


41

RACS Functionality• Admission Control: Apply admission control to resource

reservation requests– Based on knowledge of transport resource availability over the

“last-mile access” and aggregation segments of the access network

• NAPT/ Gate Control: controls near-end and far-end NAPT and FW functions, when required, between:– two core TISPAN NGN networks or,

– at the border between core and access TISPAN NGN networks


42

RACS Architecture


43

RACS Architecture• SPDF: Service-based Policy Decision Function

– Provides to AF a single point of contact

– Authorization decision for QoS resource

• A-RACF: Access Resource and Admission Control Function– Located in the access network

– Resource reservation and admission control

– In Guaranteed QoS mode, it sets L2/L3 QoS policies in RCEF

– In Relative QoS mode, it sets dynamically the diffserv QoS parameters in RCEF


44

RACS Architecture• C-BGF: Core Border Gateway Function

– Essentially, an edge router

– Located at the border of networks (access/core – core/core)

– NAPT, Gate Control, packet marking, usage metering, policing enforcement function

– SPDF controls the gates of the C-BGF based on a 5-tuple (source and destination IP addr., source and destination port numbers, protocol)

• RCEF: Resource Control Enforcement Function– Performs policy enforcement under the control of the A-RACF

– Located in the access network

– Gate control, packet marking, policing

– A-RACF controls the gates of the RCEF based on a 5-tuple (source and destination IP addr., source and destination port numbers, protocol)


45

RACS Architecture• L2TF: Layer 2 Termination Functions

– Layer 2 (e.g., PPP, ATM) is terminated here

– Authorization decision for QoS resource

• AF: Application Function– Officially, not part of RACS. Just a RACS user

– Requests bearers resources, gest informed when resources are reserved and released

– It is specific of the application. E.g., P-CSCF in IMS


46


47

NASS Functionality

• Dynamic provision of IP address and other user equipment configuration parameters (e.g., using DHCP)

• User authentication, prior or during the IP address allocation procedure

• User authentication based on user network profile– Based on PPP, IEEE 802.11X or IETF PANA


48

NASS Functionality• Line authentication based on Layer 2 line

identification

• Location management (e.g. for emergency call, …)

• Customer Premises Equipment configuration

• The NASS can be distributed between a visited and a home network– Allows nomadicity and roaming

• P-CSCF announcement


49

NASS Architecture


50

NASS Architecture• Network Address Configuration Function (NACF):

– IP address allocation to the CPE

– Distribution of other network configuration parameters such as address of DNS server(s), address of signaling proxies for specific protocols (e.g., P-CSCF)

– Typically implemented as RADIUS servers or DHCP servers

• Access Management Function (AMF):– Translates network access signaling between CPE and NACF/UAAF

– Forwards requests to he UAAF to authenticate the user, authorize/deny network access, and retrieve user-specific access configuration parameters

– Typically implemented as RADIUS client if the NACF is a RADIUS server.


51

NASS Architecture• Connectivity Session Location Repository Function (CLF):

– Registers association between the IP configuration of the CPE and

– access transport characteristics, line identifier, IP edge identity, geographical location, etc.

– Provides user network profile information to the RACS.

– Provides location information to TISPAN NGN core subsystems

• User Access Authorization Function (UAAF):

– Performs user authentication and authorization based on user profiles

– Collects accounting data


52

NASS Architecture• Profile Database Function (PDBF):

– Stores the user network profile, containing• User identity

• Supported authentication methods

• Keys

– Can be co-located with the UPSF

• Customer Premises Equipment Configuration Function (CPECF):– Provides the CPE with additional initial configuration information

(firewall, diffserv packet marking, etc.)

• Customer Network Gateway (CNG):– Single point of entrance in the customer’s network (e.g., ADSL router)

– Participates in line and access authentication


53

Service Layer• Modular Subsystem design

• Two subsystem targeted in NGN Release 1:– PSTN/ISDN Emulation Subsystem (PES)

• Emulates the PSTN/ISDN, users don’t notice the change

• Support for legacy equipment including black phones

– PSTN/ISDN Simulation Subsystem (a.k.a. IMS)• Multimedia terminals, new services based on the IMS model.

• Includes PSTN/ISDN simulation services (similar to the PSTN/ISDN Supplementary Services)

• Other subsystem may come in future NGN releases– Streaming Subsystem

– Content Broadcasting Subsystem

– Others as needed


54


55

IMS Architecture in NGN


56

NGN Impacts to IMS• IMS can be mostly reused, as specified by 3GPP,

due to its large access independence

• A few changes are required:– NAT traversal support in various points

– Relaxation of IMS constraints:• SIP compression is no longer mandatory

• SIP timers are not “wireless” specific

• SIP preconditions (183, PRACK, UPDATE) are no longer mandatory

• Support for ADSL access in P-Access-Network-Info header


57

IMS Architecture• Call Session Control Function (CSCF):

– Establishes, monitors, supports and releases multimedia functions

– Three types of CSCFs: P-CSCF, I-CSCF, S-CSCF (same as in 3GPP IMS)

– But P-CSCF contains also:• Application Level Gateway (ALG) to interface NAPT

• Interfaces the NASS to retrieve user’s physical location

• Media Gateway Control Function (MGCF):

– Controls a T-MGW

– Converts ISUP to SIP

– Functionality is equal to the 3GPP IMS MGCF


58

IMS Architecture• Multimedia Resource Function Controller (MRFC):

– Controls an MRFP

– Together with the MRFP provides conference services, announcement playback, and media transcoding

– Functionality is equal to the 3GPP IMS MRFC

• Breakout Gateway Control Function (BGCF):– Selects the network in which PSTN breakout is to occur

– Or selects the MGCF

– Equal to the 3GPP IMS BGCF


59

PSTN/ISDN Simulation Services

• ETSI TISPAN considers two types of IMS services:– Basic PSTN/ISDN Simulation Services:

• Similar (but not equal) to PSTN/ISDN Supplementary Services

• Unlike in the PSTN, not all these services require network support

• Services must interwork with the similar PSTN/ISDN supplementary service and 3GPP supplementary service


60

PSTN/ISDN Simulation Services• Multimedia services:

– New services, e.g., presence, multimedia instant messaging, SIP services that do not require extra standardization effort

• PSTN/ISDN Simulation Services defined in NGN Release 1 with three priorities:– Mandatory services

– Strictly recommended services

– Optional services


61






• ETSI

• With Softswitch

• MPLS


62

NGN Architecture

with Softswitch


63

Access Layer• Functions of access layer

– Connecting subscribers, AN & PABX and trunks from PSTN, ISDN, and PLMN, etc

– Converting the format of information (circuits to packet or packet to circuit) before transmitting it.

• Components on the access layer– IP terminals (IP phones, IP PBX and software phones, typically

intelligent terminals based on either H.323 or SIP)

– Integrated access device (IAD) (used to access subscribers (analog, ADSL, IP) in the NGN. Accesses data of subscriber terminals, voice services and video services to the packet network)


64

Access Layer– Access gateway (AGW) (acts as the interface to the core IP

network and connects subscribers with analog subscriber access, integrated services digital network (ISDN) subscriber access, PABX and x digital subscriber line (xDSL) access)

– Access network (AN) (provides connectivity between the customer premises equipment and the access gateways in the service provider's network)

– Access technologies could be cable, xDSL, wireless, optical, ...

– SIP phone (a multimedia device working in the session initiation protocol (SIP))


65

Access Layer– H. 323 phone (a multimedia device working in the

H.323 protocol)

– Signaling gateway (SG) (provides signaling interface between the IP network and the PSTN signaling network, terminates SS7 links and provides message transport part (MTP) Level 1 and Level 2 functionality)

– Trunk Media gateway (TMG) (resides between the CS network and the IP network, converts format between PCM signal flow and IP media flow, and connects devices such as PSTN exchange, PBX and BSC)


66

Access Layer– Session border controller (SBC) (located at the administrative boundary of

an IP network) for enforcing policy on multimedia sessions). The functions of SBC are

• Interworking

• Security

• Management of service level agreements

• Overload control

• NAT and firewall traversal

• lawful interception

• QoS management

• Protocol translation

• Call accounting


67

NGN Architecture

with Softswitch


68

Transport Layer• Composed of devices such as routers and Layer 3

switchers located in the backbone networks

• Packet switching

• Provides common and integrated platform for data transport

• Ensures– High reliability

– QoS assurance

– High Capacity


69

NGN Architecture

with Softswitch


70

Control Layer• Adopts software switching technologies to achieve

– Primary real-time call control

– Connection control

• Functions of softswitch– Call control

– Media gateway access control

– Signaling gateway control

– Border gateway control

– Resource allocation

– Protocol processing

– Routing

– Authentication and charging


71

NGN Architecture

with Softswitch


72

Service Layer

• The components are– Operation support system (OSS)

• Integrated charging, network operation and management

– Application server• Produces and manages value-added services and IN

services

• Provides a platform for third-party services

• Helps to develop supplementary services


73

Service Layer– Media server (MS)

• Service tone playing, conference services, IVR, recorded announcement, …

– Service control point (SCP)• Stores subscriber data and service logics

• Helps to realize various intelligent calls

– Video server• Schedules, manages and provides video conferences


74




• MPLS Basics




• LDP


75

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


76

Need for MPLS

• Forwarding function of a conventional router– A capacity demanding procedure

– Constitutes a bottle neck with increase in line speed

• MPLS simplifies forwarding function– By introducing a connection oriented mechanism

inside the connectionless IP networks


77




• MPLS Basics




• LDP


78

IPControl Plane:Connectionless Oriented

Forwarding Plane:Connectionless Oriented

Why MPLS

Control Plane:Connectionless Oriented

Forwarding Plane:Connection OrientedMPLS

Control Plane:Connection Oriented

Forwarding Plane:Connection Oriented

ATM

Weak requirement of network topologyL3 routing

Low efficiency of routingPoor QoS

Router ATM Switch MPLS Router

R X X

Layer 3 routing: Scalability and flexibilityLayer 2 routing: Reliability and QoS guarantee

L2 routingQoS guarantee provided

N2 problem exists


79

MPLS Architecture

LER

LER

LER

LER

LSR LSR

LSR

MPLS Domain

IP

MPLS

LSP

• LSR ： Label Switch Router

• LER ： Label Edge Router

• LSP ： Label Switch Path


80

MPLS Architecture

Core LSR

IP IP L1 IP L2 IP L3 IP

Traditional IP Forwarding

Label Forwarding

LER LER

Traditional IP Forwarding


81




• MPLS Basics




• LDP


82

Label Switching

• Decomposition of network layer routing into control and forwarding components

• Label switching forwarding component algorithm uses– Forwarding table

– Label carried in the packet

• What is a Label?– Short fixed length entity


83




• MPLS Basics




• LDP


84

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 Ri+1

• 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


85

MPLS Basics• 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


86

IP TTL --MPLS TTL＝ IP TTL MPLS TTL --

MPLS TTL --IP TTL＝MPLS TTL

Entrance LER LSR Exit LER

Count MPLS TTL to IP TTL

TTL of MPLS

IP TTL --MPLS TTL＝ 255 MPLS TTL -- IP TTL --

Entrance LER LSR Exit LER

One hop for the whole MPLS domain


87




• MPLS Basics




• LDP


88

MPLS Frame Format and Label

• For ATM and FR– shim mode/frame mode

– Cell mode: VC (VPI/VCI for ATM and DLCI for FR) as the label

L2 Header MPLS Header IP Header Data

Label EXP S TTL

0 20 23 24 31


89

The Location of Label

Header Lable L3 Data

Ethernet/SONET/SDH

ATM Header Lable L3 Data

Frame mode, ATM

Cell mode, ATM

VPI/VCI L3 Data


90

Label Stack

• Theoretically, the nest of label stack is unlimited to provide unlimited capacity supporting new services.

L2 Header MPLS Header

IP Header Data

MPLS Header

MPLS Header

…


91




• MPLS Basics




• LDP


92

The Concept of Label Forwarding• FEC （ Forwarding Equivalence Class ）

– Use the same LSP for packets with similar or identical characteristics (bound to the same label)

• NHLFE （ Next Hop Label Forwarding Entry ）– Includes the information of the next hop

– Action types: push/pop/swap/null

• FTN （ FEC to NHLFE ）– Mapping FEC to NHLFE

• ILM （ Incoming Label Map ）– Mapping MPLS label to NHLFE


93

Label Forwarding

• To establish a LSP: With routing protocol and LDP, build the routing table and FEC-Label mapping for FECs

• Push the Labels for different FECs

A:

…Push L1E1B10.0.1.0/24

OthersActionSend toNext Hop

NHLFEFEC

Action: pop

Exit LER

Bind LSP according to the IP HeaderFTN->NHLFE

ILM->NHLFE

ILM->NHLFEFind out the next hop according to the IP HeaderILM->NHLFE

Entrance LER LSR LSR

Action: push

Action: swap Action: swap

A B C DE1 E2 E3


94

Label Forwarding

• In the MPLS domain, forwarding is based on the Label and the forwarding table

B:

…Swap, L1->L2E2CL1


NHLFEOriginal Label

Action: pop

Exit LER


ILM->NHLFE



Action: push


A B C DE1 E2 E3


95

Label Forwarding

• Remove the Label and forward

Exit LER


ILM->NHLFE



Action: push


A B C D

D:

…Pop LabelL3


NHLFEOriginal Label

Action: pop


96

Penultimate Hop Popping

• Pop the Label at the hop before the last hop, since the Label is not useful at the last hop

• In case of only one Label, use IP routing at the last hop, otherwise forward the message according to other Labels.

Exit LER


ILM->NHLFE

Find out the next hop according to the IP HeaderILM->NHLFE


Action: push

Action: swap Action: pop

A B C D


97

Establishing LSP• LSP drivers

– Stream driven• Establish LSP when message is received

– Topology driven• Establish LSP based on the topology information (routing

information)

– Application driven • Establish LSP based on the applications

• Signaling Protocol– For Label allocation among LSRs, and establishing

LSP

– LDP, CR-LDP, RSVP-TE, MP-BGP, PIM


98




• MPLS Basics




• LDP


99

Label Distribution and Management• Label distribution modes

– DoD: downstream-on-demand

– DU: downstream unsolicited

• Label control modes– Ordered

– Independent

• Label retention modes– Conservative

– Liberal


100

Label Distribution Mode: DoD

171.68.1.0/24171.68.4.0/24

LSR1 LSR2 LSR3

Label 18 is allocated to 172.68.1.0/24

Label 20 is allocated to 172.68.1.0/24

Request the Label for dest-addr 172.68.1.0/24

Request the Label for dest-addr 172.68.1.0/24

Trigger

Sender Receiver


101

Label Distribution Mode: DU

171.68.1.0/24171.68.4.0/24

Label 18 could be used to get to 172.68.1.0/24

Trigger

Label 20 could be used to get to 172.68.1.0/24

Sender Receiver


102

Label Control Mode

Label Request

Label Request

Label Request

Label Mapping

Label Mapping

Label Mapping

Sender Receiver


103

Label Retention Mode: Conservative• Only keep the Label from the next hop neighbor,

while discarding the other labels– Advantages: Less memory space required

– Disadvantages: Slow convergence with changing topology

LSR1 LSR2 LSR3 LSR4

LSR5

172.16.2/24

mappinglabel 20

mappinglabel 30

mappinglabel 17

mappinglabel 16

Not the label from the next hop neighbor, discard it


104

Label Retention Mode: Liberal• Keep the labels from any neighbor

– Advantage: Fast convergence with changing topology

– Disadvantages: More memory space required

LSR1 LSR2 LSR3 LSR4

LSR5

172.16.2/24

mappinglabel 20

mappinglabel 30

mappinglabel 17

mappinglabel 16

Not the label from the next hop neighbor, keep it for future use


105




• MPLS Basics




• LDP


106

LSP Loop Detection

• Loop routing– Very harmful

• Two ways to prevent loop routing– Max hop

– Path vector


107




• MPLS Basics




• LDP


108

Concept of LDP• Label Distribution Protocol

– To generate dynamic labels, similar to dynamic routing protocols (e.g., OSPF)• With messages

• With neighbor discovery and maintenance

• With the algorithm to obtain the results based on the information

• The result of LDP is composed of labels, but the result of OSPF is routing information

– Main functions• Distribute Label-FEC mapping

• Establish and maintain the routing for label switch

• LDP uses UDP and TCP


109

Messages in LDP• Discovery message

– For informing and maintaining LSR

• Session message– For establishing, maintaining and stopping the session

connection between LDP entities

• Advertisement message– For creating, changing and removing FEC-Label

binding

• Notification message– For notification of information and error


110

LDP Session Establishment and MaintenanceNeighbor Discovery: Sending hello messages (UDP/prot:646）

TCP Connection Establishment: Initiated by the LSR with higher address (TCP/port:646)

Session Initialization: Initialization message sent by Master, with negotiation parameters.

Slave checks if the parameters are acceptable. If so, the initialization message with the negotiation parameters, and the keepalive message are sent

Master checks if the parameters are acceptable, If so, the keepalive message is sent

By receiving the keepalive message by both LSRs, the session is established

Disconnect the TCP connection and stop the session at any error message

M

M

M

M

M

LSR1 LSR2

S

S

S

S

S


111

A Brief Review• Background of NGN

– Multi-service

– Mature of new technologies

– Flexibility

• Definition of NGN– By ITU-T

• Characteristics of NGN

• Architectures of NGN– With IMS by 3GPP

– By ETSI

– With Softswitch


112

A Brief Review• MPLS

– Simplifies forwarding function

– Integrates ATM and IP

• Label switching– Decomposition of network layer routing into control and

forwarding components

• Label forwarding

• Label distribution and management– Label distribution, control and retention

第二章 下一代网络与 mpls chapter 2 next generation network and mpls

Documents

第二章下一代网络与 mpls chapter 2 next generation network and mpls