MPLS-自己的经验理解通俗易懂.ppt

《MPLS-自己的经验理解通俗易懂.ppt》由会员分享，可在线阅读，更多相关《MPLS-自己的经验理解通俗易懂.ppt（94页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、MPLS及MPLS VPN基本概念,2012年3月,MPLS VPN的基本概念,目录,MPLS的基本概念,1,3,MPLS 及MPLS VPN举例,传统IP路由网络的缺陷,传统的IP数据转发,使用路由协议传送IP路由信息 基于IP包的目标地址进行数据转发 IP包每经过一个路由器都需要进行路由表的查询,IP的逐跳转发，在经过的每一跳处，必须进行路由表的最长匹配查找（可能多次），速度缓慢。,在传统的IP转发中的流量工程问题,Most traffic goes between large sites A and B and uses only the primary link. Destinatio

2、n-based routing does not provide any mechanism for load balancing across unequal paths. Policy-based routing can be used to forward packets based on other parameters, but this is not a scalable solution.,Primary OC192 link,Large Site A,Large Site B,Small Site C,BackupOC48 link,Review Questions,列出主要的

3、传统IP路由缺点. IP包的传发是基于那一种信息? 为什么这种转发机制不适用于大型网络?,MPLS架构及相关技术,MPLS数据转发,MPLS的标签转发，通过事先分配好的标签，为报文建立了一条标签转发通道（LSP），在通道经过的每一台设备处，只需要进行快速的标签交换即可（一次查找）。,MPLS:多协议标签交换,MPLS:Multi-Protocol Label Switching 在IP网络实现2.5层数据交换,MPLS 的基本概念,基于标签进行数据转发的机制 标签对应于IP目标路由网络 标签可对应于其他相关参数 Qos IP源地址 支持多种协议的转发,MPLS/IP网络,MPLS 架构,控制层

4、面（Control plane） 运用路由协议进行路由信息的交换 运用标签分发协议进行标签交换 数据层面（Data plane） 基于标签进行数据转发,MPLS Architecture,Router functionality is divided into two major parts: control plane and data plane,Data Plane,Control Plane,OSPF: 10.0.0.0/8,LDP: 10.0.0.0/8 Label 17,OSPF,LDP,LFIB,LDP: 10.0.0.0/8 Label 4,OSPF: 10.0.0.0/8,L

5、abeled packet Label 4,Labeled packet Label 17,Label Format,MPLS uses a 32-bit label field that contains the following information: 20-bit label 3-bit experimental field 1-bit bottom-of-stack indicator 8-bit time-to-live (TTL) field,LABEL,EXP,S,TTL,0,19,22,23,31,20,24,Frame-Mode MPLS,Routing lookup a

6、nd label assignment,Label Switch Router,Label switch router (LSR) 转发打了标签的IP包 Edge LSR 给IP包打标签并转发到MPLS域 删除标签并把IP包从MPLS域转发出去,MPLS Domain,Edge LSR,LSR,10.1.1.1,L=3,L=5,L=43,L=31,20.1.1.1,10.1.1.1,20.1.1.1,LSR的功能架构,LSRs, regardless of the type, perform the following three functions: Exchange routing inf

7、ormation Exchange labels Forward packets (LSRs and edge LSRs) The first two functions are part of the control plane. The last function is part of the data plane.,Architecture of LSRs,LSRs primarily forward labeled packets.,LSR,Control Plane,Data Plane,Routing Protocol,Label Distribution Protocol,Lab

8、el Forwarding Table,IP Routing Table,Exchange of routing information,Exchange of labels,Incoming labeled packets,Outgoing labeled packets,Architecture of Edge LSRs,Edge LSR,Control Plane,Data Plane,Routing Protocol,Label Distribution Protocol,Label Forwarding Table,IP Routing Table,Exchange of routi

9、ng information,Exchange of labels,Incoming labeled packets,Outgoing labeled packets,IP Forwarding Table,Incoming IP packets,Outgoing IP packets,MPLS 转发,LSR功能: 插入（Insert）标签 交换（Swap）标签 删除（Pop）标签,MPLS 域,MPLS Forwarding (Frame-Mode),On ingress a label is assigned and imposed by the IP routing process. L

10、SRs in the core swap labels based on the contents of the label forwarding table. On egress the label is removed and a routing lookup is used to forward the packet.,10.1.1.1,MPLS 网络IP路由示例,LSR,Control Plane,Data Plane,OSPF:,RT:,LIB:,FIB:,LFIB:,OSPF: 10.0.0.0/8,10.0.0.0/8 1.2.3.4,10.0.0.0/8 1.2.3.4,LSR

11、,Control Plane,Data Plane,OSPF:,RT:,LIB:,FIB:,LFIB:,OSPF: 10.0.0.0/8,10.0.0.0/8 1.2.3.4,10.0.0.0/8 1.2.3.4,LDP: 10.0.0.0/8, L=3,10.0.0.0/8 Next-hop L=3, Local L=5,LDP: 10.0.0.0/8, L=5,MPLS 网络IP路由示例,标签的分配和分发过程,IP路由协议构造IP路由表 LSR对路由表中每一目标网段独立地分配标签 LSR把所分配的标签公告给其他LSR 根据所受到的标签，LSR构建LIB，LFIB和FIB,路由表的构建,IP

12、 routing protocols are used to build IP routing tables on all LSRs. FIBs are built based on IP routing tables with no labeling information.,A,B,C,D,E,Network X,分配标签,Every LSR allocates a label for every destination in the IP routing table. Labels have local significance. Label allocations are asynch

13、ronous.,A,B,C,D,E,Network X,LIB 和 LFIB 的建立,LIB and LFIB structures have to be initialized on the LSR allocating the label.,A,B,C,D,E,Network X,标签分发 Label Distribution,The allocated label is advertised to all neighbor LSRs, regardless of whether the neighbors are upstream or downstream LSRs for the d

14、estination.,标签通告的接收（Receiving Label Advertisement）,Every LSR stores the received label in its LIB. Edge LSRs that receive the label from their next-hop also store the label information in the FIB.,X = 25,X = 25,A,B,C,D,E,X = 25,Network X,过渡期的数据传送（Interim Packet Propagation）,Forwarded IP packets are

15、labeled only on the path segments where the labels have already been assigned.,IP: X,Lab: 25,IP: X,A,B,C,E,进一步的标签分配（Further Label Allocation）,Every LSR will eventually assign a label for every destination.,A,B,C,D,E,Network X,标签通告的接收（Receiving Label Advertisement）,Every LSR stores received informati

16、on in its LIB. LSRs that receive their label from their next-hop LSR will also populate the IP forwarding table (FIB).,A,B,C,D,E,Network X,增加LFIB条目（Populating LFIB）,Router B has already assigned a label to X and created an entry in the LFIB. The outgoing label is inserted in the LFIB after the label

17、 is received from the next-hop LSR.,Label,Action,Next hop,25,47,C,LFIB on B,A,B,C,D,E,Network X,数据包通过MPLS网络的过程,IP: X,IP: X,A,B,C,E,Lab: 25,Lab: 47,MPLS网络LSP的建立,MPLS网络的优化,MPLS Domain,Double lookup is not an optimal way of forwarding labeled packets. A label can be removed one hop earlier.,10.0.0.0/8

18、L=19,10.0.0.0/8 L=18,10.0.0.0/8 L=17,Double lookup is needed: 1.LFIB: remove the label. 2.FIB: forward the IP packet based on IP next-hop address.,10.0.0.0/8,倒数第二跳弹出（Penultimate Hop Popping）,MPLS Domain,A label is removed on the router before the last hop within an MPLS domain.,10.0.0.0/8 L=pop,10.0

19、.0.0/8 L=18,10.0.0.0/8 L=17,10.0.0.0/8,Pop or implicit null label is advertised.,小结,MPLS VPN的基本概念,目录,MPLS的基本概念,1,3,MPLS 及MPLS VPN举例,什么是VPN?,Customer Site,Large Customer Site,VPN术语（VPN Terminology）,用户网络(C-network): the part of the network still under customer control,运营商网络(P-network): the service pro

20、vider infrastructure used to provide VPN services,用户站点: a contiguous part of the customer network (can encompass many physical locations),VPN业务网络视图,VPN的分类类型,Overlay VPN（一层VPN）,运营商提供物理层的连接 用户负责数据链路层和ip层 用户自行管理路由,ISDN,E1, T1, DS0,SDH, SONET,PPP,HDLC,IP,Overlay VPN（二层VPN）,运营商提供数据链路层的连接 用户负责ip层 用户自行管理路由

21、,X.25,Frame Relay,ATM,IP,Overlay VPN（IP隧道）,用户负责ip层 用户自行管理路由,Generic Route Encapsulation (GRE),IP Security (IPSec),IP,IP,Service Provider Network,Peer-to-Peer VPN Concept,Customer Site,Router A,Customer Site,Router B,Customer Site,Router C,Customer Site,Router D,PE,Router,PE Router,PE Router,PE Route

22、r,Routing information is exchanged between CE and PE routers.,PE routers exchange customer routes through the core network.,Finally, the customer routes propagated through the PE network are sent to other CE routers.,共享PE的方式,专用PE的方式,MPLS VPN,路由型MPLS VPN的架构,客户边界路由器,运营商边界路由器,运营商路由器,VPN路由及转发表（VRF）,PE的路

23、由表,地址复用,路由区分器（Route Distinguisher）,RD：64比特地址 用于区分PE中每个用户的路由 VPNv4地址=RD+IPv4地址 VPNv4地址通过BGP在PE之间进行交换 多协议BGP（MP-BGP）,路由区分器的运用,使用路由区分器,路由标记（Route Targets）,多个用户站点分属于不同的VPN，需要使用RT标记各自的VPN路由 附加在VPNv4路由中传送以标记不同的VPN RT加入到BGP的扩展属性中进行传送 RT的灵活应用可支持不同的VPN拓扑,RT的工作原理,Export RT：路由发送标记，定义VPN组 Import RT：路由接收标记，识别VPN

24、组 在发生端的PE，IPv4转换成VPNv4路由时加入Export RT 在接收端的PE，根据Import RT进行检查 收到的路由的RT与Import RT匹配，接收路由,RT的灵活应用1,RT的灵活应用2,RT的灵活应用3,路由型MPLS VPN的路由模型,MPLS VPN路由,CE运行路由协议 PE运行路由协议与CE交换路由信息 PE运行MPLS传送VPN路由 P运行MPLS,CE,PE,PE路由器的路由,P,MPLS VPN端到端的路由信息流1,MPLS VPN端到端的路由信息流2,MPLS VPN端到端的路由信息流3,路由型MPLS VPN的数据转发,传送原始IP数据包,传送打了标签

25、的IP包,给IP包打两次标签,VPN标签由Ingress PE路由器标记并发布,MPLS L2VPN,MPLS L2VPN,MPLS L2VPN 提供基于 MPLS网络的二层 VPN服务， 使运营商可以在统一的 MPLS 网络上提供基于不同数据链路层的二层 VPN。 简单来说，MPLS L2VPN 就是在 MPLS 网络上透明传输用户二层数据。从用户的角度来看，MPLS网络是一个二层交换网络，可以在不同节点间建立二层连接。 相对于 MPLS L3VPN，MPLS L2VPN 具有以下优点： 可扩展性强：MPLS L2VPN 只建立二层连接关系，不引入和管理用户的路由信息。 可靠性和私网路由的安

26、全性得到保证 支持多种网络层协议：包括 IP、IPX等,MPLS L2VPN的基本概念,在 MPLS L2VPN 中，CE、PE、P 的概念与 MPLS L3VPN 一样，原理也相似。 MPLS L2VPN 通过标签栈实现用户报文在 MPLS 网络中的透明传送： 外层标签（称为 Tunnel 标签）用于将报文从一个 PE 传递到另一个 PE； 内层标签（称为 VC 标签）用于区分不同 VPN 中的不同连接； 接收方 PE 根据 VC 标签决定将报文转发给哪个 CE。,MPLS L2VPN 标签栈处理,MPLS L2VPN 的实现方式,还没有形成正式的标准。IETF 的 PPVPN工作组制订了多

27、个框架草案，其中最主要的两种称为 Martini 草案和 Kompella 草案： draft-martini-l2circuit-trans-mpls draft-kompella-ppvpn-l2vpn Martini 草案定义了通过建立点到点的链路来实现 MPLS L2VPN 的方法。它以 LDP为信令协议来传递双方的 VC 标签，称为 Martini 方式 MPLS L2VPN。 Kompella 草案则定义了在 MPLS 网络上以端到端（CE 到 CE）的方式建立 MPLS L2VPN。目前它采用扩展了的 BGP为信令协议来发布二层可达信息和 VC 标签，称为 Kompella 方式

28、 MPLS L2VPN。,MPLS VPN的基本概念,目录,MPLS的基本概念,1,3,MPLS 及MPLS VPN举例,衢州电信城域网,MPLS 域,衢州电信城域网核心网,MPLS 域,LSR,Edge LSRs,衢州电信城域网MPLS VPN环境,MPLS 环境,P,PE,城域网三层MPLS VPN实例（环保监控）,江山SR1： description CTVPN45002-HuangBaoJianKong vrf-import vprn200017_import route-distinguisher 4809:45002 auto-bind ldp vrf-target target:

29、4809:4500200 interface ge-lag-2.3899 create description HBJK_HuangBaoJu address 42.10.254.25/30 local-proxy-arp sap lag-2:3899.* create ingress qos 105 exit egress qos 400 exit exit exit interface ge-lag-2.3910 create description HBJK_HengChangShiYe address 42.10.41.161/28 local-proxy-arp sap lag-2:

30、3910.* create ingress qos 105 exit egress qos 400 exit exit exit,龙游SR1: description CTVPN45002-HuangBaoJianKong vrf-import vprn200017_import route-distinguisher 4809:45002 auto-bind ldp vrf-target target:4809:4500200 interface ge-5/1/2.3901 create description HBJK_TianTingYaLun address 42.10.33.17/2

31、8 sap 5/1/2:1592.3901 create ingress qos 105 multipoint-shared exit egress qos 400 exit exit exit interface ge-5/1/2.3907 create description HBJK_JuHuaKuangYe address 42.10.33.113/28 sap lag-2:3907.* create ingress qos 105 exit egress qos 400 exit exit exit,城域网二层MPLS VPN实例（邮政）,南区SR1： description VPL

32、S_Youzheng stp shutdown exit sap 5/1/1:3800.* create ingress qos 111 exit egress qos 210 exit exit sap 5/1/8:1304.3800 create description HeHuaSanLu ingress qos 111 exit egress qos 210 exit exit mesh-sdp 128:303800 create exit mesh-sdp 131:303800 create exit,江山SR1: split-horizon-group 303800 create exit stp shutdown exit sap lag-1:1452.1357 split-horizon-group 303800 create exit sap lag-1:1443.3800 split-horizon-group 303800 create exit sap lag-1:3800.* create exit mesh-sdp 130:303800 create exit,网络操作维护中心2012年3月,