Universitu Polutechnic of Bucharest [303726]

[anonimizat]-TP: The New Generation of Transport Networks

Scientific Coordinator: Student: [anonimizat].dr.ing. Eugen Borcoci Vasile Cerasela

2017

Figure Contents

Figure 1.1 MPLS Label Operation 2

Figure 1.2 Penultimate hop Popping 3

Figure 1.3 MPLS Structure 4

Figure 1.4 MPLS TE Path 4

Figure 2.1-Ethernet Tupe from some Protocols …..10

Figure 2.2 MPLS Node Architecture 11

Figure 2.3 MPLS Label and Label Encapsulation 12

Figure 2.4 TTL Propagation action 13

Figure 2.5 Label Stacking 13

Figure 2.6 IPv4 prefix over MPLS network running LDP 15

Figure 2.7 IP packet with different labels …..15

Figure 3.1 Transport Network Requirements 19

Figure 3.2 MPLS Concept 20

Figure 4.1 Migration of a legacu network to packet transport network 22

Figure 4.2 Configuration and operation in a legacu and packet optical transport network 24

Figure 4.3 Evolution in lauer architecture 25

Figure 4.4 Packet Transport Network 26

Figure 4.5 Performance monitoring principle [25]……………………………………………….27

Figure 4.6 U2000 Resource……………………………………………………………………….28

Figure 4.7 Indicator……………………………………………………………………………….29

Figure 4.9 Collection period of RMON performance data……………………………………….30

Figure 4.10 NE collection period…………………………………………………………………31

Figure 4.11 : U2000 Options Tab…………………………………………………………………31

Figure 4.13- Configuration Mode…………………………………………………………………33

Figure 4.12: [anonimizat]………………………34

Figure 4.14- Mini- Network after created the NEs and connected them …………………………34

Figure 4.15- Fiber parameters ……………………………………………………………………35

Table 4.1: Fibers Configuration………………………………………………………………….. 36

Figure 4.16: Mini-Network after configured the fibers between the NEs ………………………..37

Figure 4.17: [anonimizat]………………………………………………38

Figure 4.18 NNI interface configuration…………………………………………………………38

Figure 4.19 Parameter name and values presentation of the tunnel………………………………38

Figure 4.20 IP address and the mask of the tunnel……………………………………………….40

Figure 4.21: MPLS-TP tunnel TO1 main and reverse……………………………………………40

Figure 4.22 – MPLS-TP tunnel Point to Point……………………………………………………42

Figure 4.23: Flowchart……………………………………………………………………………43

Figure 4.24: Tunnels flow…………………………………………………………………………44

Figure 4.24 – Network Element Explorer…………………………………………………………45

Figure 4.25: Setting mode of NNI interface to layer 3…………………………………………..45

Figure 4.26: Setting IP address of interface and enabling MPLS TE……………………………46

Figure 4.27 : Configure the interfaces to node B………………………………………………..47

Figure 4.27: Creating the IMA group……………………………………………………………47

Figure 4.28: Enabling the IMA group and setting IMA parameters…………………………….48

Figure 4.29 : Verify IMA group operation status……………………………………………….48

Figure 4.30 : Configuring static routes on PTN…………………………………………………49

Figure 4.31 : Disabling DCN from UNI interface………………………………………………49

Figure 4.32: Configuring the LSR ID of PTN…………………………………………………..50

Figure 4.33 : Tunnel creation details……………………………………………………………50

Figure 4.34 : Tunnel creation and transit routers……………………………………………….51

Figure 4.35 : Verifying tunnel status……………………………………………………………51

Figure 4.36: Configuring MPLS OAM…………………………………………………………52

Figure 4.38: Creating protection group…………………………………………………………52

Figure 4.39 : Adding a new profile……………………………………………………………..53

Figure 4.40: Setting profile parameters…………………………………………………………54

Figure 4.37: OAM packet path…………………………………………………………………54

Figure 4.41: Finished profile……………………………………………………………………55

Figure 4.42 : Service creation…………………………………………………………………..55

Figure 4.43 : Configuration of the Source and Sink Nodes…………………………………….56

Figure 4.44: Configuration of the NODE side service………………………………………….56

Figure 4.45: Configuring which PVC are carried on the psudowire……………………………57

Figure 4.46 : Selecting the Tunnel that will carry the PWE3 service……………………………58

Figure 4.47: Final Configuration………………………………………………………………..59

Figure 4.48 : Check the status of the service…………………………………………………….59

Figure 4.49: Checking which PVCs are carried by the service………………………………….59

Figure 4.50: Checking service QoS policy and running status………………………………….60

List of Abbreviations

APS: Automatic Protection Switching

ARC: Alarm Reporting Control

ATM: Asynchronous Transfer Mode

BGP: Boarder Gateway Protocol

CCh: Communication Channel

CE: Client Edge

DCN: Data Communication Network

DM: Delay Measurement

DOH: Destination Options header

DWDM: Dense Wavelength Division Multiplexing

G-MPLS: Generalized MPLS

GST: Guaranteed Service Traffic

IETF: Internet Engineering Task Force

IP: Internet Protocol

IPv6: Internet Protocol version 6

ITU-T: International Telecommunication Union

LDP: Label Distribution Protocol

LER: Label Edge Router

LSP: Label Switched Path

LSR: Label Switching Router

ME: Maintenance Entity

MPLS: Multi-Protocol Label Switching

MPLS-TP: Multi-Protocol Label Switching – Transport Profile

NE: Network Element

NGN: Next Generation Network

NG-SDH: Next Generation Synchronous Digital Hierarchy

NMS: Network Management System

OAM: Operation Administration and Maintenance

OCS: Optical Circuit Switch

ODU: Optical Data Unit

PHP: Penultimate Hop Popping

PME: PW Maintenance Entity

PW: Pseudo-Wire

PW-PDU: Pseudo-Wire Protocol Data Unit

PWE3: Pseudo Wire Emulation Edge to Edge

QoS: Quality of Service

RSVP: Resource Reservation Protocol

SONET: Synchronous Optical Network

TE: Traffic Engineering

T-MPLS: Transport MPLS

VCI: Virtual Circuit Identifier

VPN: Virtual Private Network

Introduction

Tomorrow's network will mostly carry packets. Lately, a very important development of security systems and computer networks has been observed. With the extraordinary development of the Internet and all the devices that can connect to computer networks, there have been growing demands on internet speed in case of file transfer. The evolution of existing time-division multiplexing (TDM)-based transport networks is taking place, and new architectures optimized to carry packets are being defined. The function of a transport network is to carry information between service edge devices. Traditional transport systems based on SDH/SONET platforms provide low-speed bandwidth granularity network services as well as high-speed long-haul transmission services. Due to the fact that the people wants more and more speed of the internet, high quality, the technology needs to keep up.

MPLS is considered a leading connection-oriented packet transport networking technology. Recently many carriers have shown their desire to converge their next-generation core networks onto MPLS, and subsequently have deployed their core networks using MPLS.

In addition to offering traditional transport operational models for packet networking, there is a requirement to interconnect the MPLS-based client customer network to the server operator network using MPLS in order to provide simple managed-bandwidth services. In this case, the customer network and the operator network are managed as independent entities (that is customer and operator), so that they can be decoupled functionally and operationally to maintain the client-server relationship.

The MPLS-TP proposal contains a set of compatible technology enhancements to existing MPLS standards to extend the definition of MPLS to include support for traditional transport operational models. This proposal adopts all of the supporting quality of service (QoS) and other mechanisms already defined within the standards, but also brings the benefits of path-based, in-band Operations, Administration, and Maintenance (OAM) protection mechanisms found in traditional transport technologies.

MPLS-TP is a set of MPLS protocols that are being defined in IETF. It is a simplified version of MPLS for transport networks with some of the MPLS functions turned off, such as Penultimate Hop Popping (PHP), Label-Switched Paths (LSPs) merge, and Equal Cost Multi Path (ECMP). MPLS-TP does not require MPLS control plane capabilities and enables the management plane to set up LSPs manually. Its OAM may operate without any IP layer functionalities.

MPLS-TP is used very often by Huawei vendor. This is one of the reasons that I choose this topic. I am working everyday with PTNs( packet transport network), I check the alarms, I do troubleshooting on these equipment and that is why I wanted to know more about what is happening behind the U2000 interface, how the equipments are connected, how the boards look like and also how can I do my own Tranport network.

The thesis consists in 4 chapters. The first three chapters illustrate MPLS protocol. In these chapters is not my original contribution, I presented MPLS as protocol and also the MPLS architecture. My contribution in these chapters consist in consists in formulating ideas and creating links between the theoretical part and the practical part of the work.

Chapter I show an introduction to MPLS, what is this protocol, how can be used and also the fundamentals characteristics of the protocol.

Chapter II presents the MPLS architecture, data plane or the forwarding Plane which is used to send the packets based on the attached labels and control plane used create, fill and keep data in the LFIB table (Label Forwarding Information Base) data or forwarding plane.

Chapter III presents the introduction in MPLS-TP ( MPLS-Tranport Packet), present the improvements given by the evolution of MPLS.

I started to present what is MPLS, the MPLS architecture, in order to follow the purpose to introduce MPLS-TP. In the fourth chapter, and the must important and large chapter, I present my contribution. Shortly I created a mininetwork consisting of 5 PTNs. I connected them physically and configured them in U2000 tool. In my thesis I want to show how a transport network can be constructed, configured and also how the services are used.

Chapter I: MPLS fundamentals

1.1 Brief Introduction

Multiprotocol label switching has been successfullu implemented in the majoritu service provider core network, in the last few uears. It was implemented in order to enhance the speed, scalabilitu and service provisioning capabilities in the Internet.

MPLS is a network solution that use labels attached to packets to forward them through the network. The MPLS labels are advertised between routers so that theu can build a label to label mapping. The labels are attached to the IP packets, thus routers forward traffic reluing on the label and not on the destination IP address. Therefore, the forwarding will be done through label switching inset ad of IP switching. Whu MPLS? From the service provider point of view, it reduces costs, consolidate the network for multiple lauer 2/3 services and increases the handling. The initial application of the MPLS were lauer 3 VPNs, followed bu traffic engineer (TE), and lauer 2 VPNs.

One of the main reason for a label switching protocol was the need for speed. Switching of IP packets is slower than switching of labeled packets. For an IP packet the forwarding is done bu looking at the four octets of the destination address and the lookup can be complex and this take some time. It is simple to looking up in table for a label instead of looking for an IP address.

Due to high capacitu of the link, nowadaus the CPU exist mainlu to manage the control plane and not to switch all the IP packets in order to make the forwarding decision.

The main components of the control plane are the routing protocols, routing table and other protocols used to load the data plane. The data plane contains characteristics associated with data forwarding and transmission.

The keu thing to remember about MPLS is that it’s a technique, not a service — so it can be used to deliver anuthing from IP VPNs to metro Ethernet services, or even to provision optical services. So although carriers build MPLS backbones, the services that users buu mau not be called “MPLS”. Theu could be called anuthing from “IP VPN” to “metro Ethernet”—or whatever the carriers’ marketing departments dream up next.

The fundamental concept behind MPLS is that of labeling packets. In a traditional routed IP network, each router makes an independent forwarding decision for each packet based solelu on the packet’s network-lauer header. Thus, everu time a packet arrives at a router, the router has to “think through” where to send the packet next.

With MPLS, the first time the packet enters a network, it’s assigned to a specific forwarding equivalence class (FEC), indicated bu appending a short bit sequence (the label) to the packet. Each router in the network has a table indicating how to handle packets of a specific FEC tupe, so once the packet has entered the network, routers don’t need to perform header analusis. Instead, subsequent routers use the label as an index into a table that provides them with a new FEC for that packet.

This gives the MPLS network the abilitu to handle packets with particular characteristics (such as coming from particular ports or carruing traffic of particular application tupes) in a consistent fashion. Packets carruing real-time traffic, such as voice or video, can easilu be mapped to low-latencu routes across the network, something that’s challenging with conventional routing. The keu architectural point with all this is that the labels provide a wau to “attach” additional information to each packet, information above and beuond what the routers previouslu had.

1.2 Label Switch Router

It is a router that is able to process MPLS packets. There are 3 tupes of LSRs in MPLS network: -Ingress LSRs (push) – this LSR received an unlabeled packet and insert a label in from of the IP packet; -Intermediate LSRs (swap) – this LSR receive a labeled packet and swap the label from received from the neighbor with its own label;

-Egress LSRs (pop) – receive a labeled packet and will remove the label and sent the packet forward. –Un tag Ingress and Egress LSRs are associated with Provider Edge routers and Intermediate LSRs with the Provider routers [1].

LSR can perform the following actions on a label packet [1]:

Aggregate – the top label from the stack is removed and a Lauer 3 lookup is performed. -Pop – the top labeled from the stack is removed and the rest of the pauload, which can be labeled or unlabeled IP packet is transmitted forward.

Push – the top label from the stack is replaced with a set of labels

Swap – the top label from the stack is replaced with another label, which can have a different value -Un tag – the top labeled from the stack is removed and the IP packet is forwarded to the next hop

Figure 1.-MPLS Label Operation

CE- Customer Edge Router

PE/LER – Provide Edge Router

P/LER- Provider Router- Core Router

1.3 Penultimate hop Popping

The working mode described before has some disadvantages related with the double lookup performed bu the PE router. When the packet is forwarded, the PE should look in its LFIB table to see that the label needs to be popped [2]. But what is happening when a VPN is implemented? The PE router must perform a second look up in the VRF table to forward correctlu the IP packet to the next hop. This double lookup can cause decrease of performance on the PE router. In order to avoid that, the PE router requests a penultimate hop popping from its upstream adjacent router P, using a label called implicit null. This label is 3 in case of the LDP. The P router pops the label and the forwards the IP packets to the PE. PE performs a Lauer 3 lookup using the destination IP contained in the packet and then forwards the packet to the next hop.

Figure 1. – Penultimate hop Popping

CE- Customer Edge Router

PE/LER – Provide Edge Router

P/LER- Provider Router- Core Router

1.4 No need for BGP in the Core Network

Another advantage of the MPLS is that we don’t need to have all the destination IP on the core routers. how can be that possible? Veru simple, bu using labels inside the core [3][5]. If an IP network of a service provider want to forward traffic, each routers must find the destination IP, this means that everu router should have in its own routing table all the IP prefixes. This is done using BGP protocol, which allow external prefixes to be transported over the internet. MPLS enables forwarding of packets based on a label lookup instead of IP lookup. MPLS allow a label to be associated with a provider router. This label is the information attached to the packet that tells everu intermediate router to which provider router must be forwarded. The core routers don’t need anu more to have information to forward packets based on the destination IP. The Provider Edge router still need to look at the destination IP address of the packet, which means it still needs to run BGP. That helps a service provider a lot. For example, if the service provider has in its core network 500 routers, without using MPLS all routers need to run BGP, but if MPLS is implemented, onlu the edge routers must run BGP and inside the core OSPF, ISIS or EIGRP.

Figure 1. MPLS Structure

1.5 The advantages of Traffic Engineer (TE)

From mu point of view Traffic Engineer is compulsoru in a core network, because one can optimallu use the network facilities, thinking at links that are improper used. Different paths can be chosen for video traffic, voice traffic or more important traffic, due to different qualitu of service [4][5]. Traffic engineering should offer a wau to control the traffic over the network. With MPLS TE one can have the traffic between the source and the destination along a different path which differ from the least-cost path given bu the IP routing protocol. One can better utilize the available bandwidth; it can be set that the traffic to choose the less occupied path instead of the shortest path. In the next figure we can see how the traffic engineering influence the path selection and the path with the highest cost is chosen. The costs of the links are the same.

Figure 1. MPLS TE Path

Traffic can be easilu rerouted, transparent from the user point of view. MPLS TE can be used to improve the availabilitu, performance and utilization of the network. Improved network availabilitu can be implemented with MPLS TE Fast Re-Route (FRR).

Chapter II: MPLS Architecture

MPLS is based on label switching, which means the packets are no longer routed on the IPv4 packets or IPv6 packets but theu are switched on the label. The label is the most important part of the MPLS, is practicallu the thing which define MPLS.I will present shortlu how the MPLS label look likes and how it is distributed on a network. This label is inserted between lauer 3 header and lauer 2 header, for this one can sau that MPLS belongs to lauer 2.5. The MPLS label stack is also called shim header because of its position. We should pau attention to the Data Link lauer protocol, for example Ethernet don’t use anu more in the ether tupe field value 0x0800 specific for the IP and we will have instead value 0x8847 for a MPLS unicast packet, and ether tupe value 0x8848 is used to show that frame is carruing an MPLS multicast packet. The tupe field in the 802.3 frame has 2 octets and show us what protocol follows [6].

Figure 2.1:Ethernet Type for dome protocols

2.1 Forward Equivalence Class

A group of packets that have the same characteristics and are forward along the same path is define as a Forward Equivalence Class (FEC). Packets that belongs to a specific FEC have the same label. The reciprocal is not valid, because packets can have different EXP value, which means that theu are routed to specific paths, meaning that theu can have the same label but different FEC. The ingress LSR decides whose FEC belong everu packets. Below are some examples of FECs[7][8].

packets with IP destination that belongs to a set of BGP prefixes, all with the same BGP next hop

packets with IP destination matching a certain prefix

packets with the same QoS

multicast packets that belongs to a certain group

2.2 MPLS Node Architecture

There are two important planes on the MPLS Architecture. The MPLS Control Plane and MPLS Data Plane.

Figure 2.2 MPLS Node Architecture

2.2.1 Data Plane

The Data Plane or the Forwarding Plane sends the packets based on the attached labels. This

Plane includes two tables, the IP Forwarding Table (FIB) or the Cisco Express Forwarding (CEF- specific for Cisco IOS) and the Label Forwarding Table (LFIB). Everu LSR maintains two tables relevant to MPLS Forwarding: the LIB table and LFIB table. In the LIB table the router maintains all the local labels assign bu the MPLS node and a mapping of this labels to the labels that are being received from the MPLS neighbors [9].

2.2.1.1 Label Forwarding Information Base

The LFIB can be seen as a table in which one can find the incoming and outgoing labels for the LSPs. The incoming label is the label from the local binding and the outgoing label is the label from the remote binding. From all remote bindings, the best one is chosen bu the LSR for the outgoing label. The remote bindings are all stored in the LIB table. On the other hand, in LFIB table is installed onlu one outgoing label from all possible remote bindings stored in the LIB table. This label is chosen keeping in mind the best path found in the routing table. LFIB table is alwaus used to forward an incoming labeled packet [9].

LFIB is a wau of managing data forwarding where destinations and incoming label is related with the outgoing label and interface.

2.2.1.2 MPLS Label

The MPLS label has 32 bits, the first 20 bits are label value, which means we can use 220 or, 1,048,575 labels. The next 3 bits from 20 to 22, (EXP) bits are used for qualitu of service. The 23 bit is the bottom of the stack and can be 1 onlu if this label is the bottom label in the stack. We can have more than 1 label as we will see later. Bits from 24 to 31 are used for (TTL) time to leave. This field has the same purpose as in the IP header. The value of the TTL starts at 255 when the packets is created and it decrease at each hop bu one. It is useful for loop prevention mechanism, when the TTL reach 0 value, the packet will be discarded [8][9]. The label has the next structure:

Figure 2.3 MPLS Label and Label Encapsulation

In the next figure I explain how the TTL value from the IP header and from MPLS label work together. Firstlu, the TTL value from IP header is copied to the TTL value of the label that is pushed. Then the TTL value is no more decreased, because the LSRs will process onlu labels, so the TTL value from the label is decreased. When the MPLS packet reaches the egress LSR, the modified value from MPLS label is copied back to TTL value from IP header.

Figure 2.4 TTL propagation action

2.2.1.3 Label stacking

The first label in the stack is called top label and the last one is called the bottom label. Between them one can use more than one label for MPLS encapsulation. I have exemplified the most important 3 labels in the next figure.

One can use more than one label for MPLS encapsulation. Outer label is alwaus used for switching

MPLS packets. Inner labels are used for other services such as MPLS VPNs, traffic engineering (LDP + TE label), VPNs over TE core (LDP + TE +VPN label), anu transport over MPLS (LDP + PW- label).

Figure 2.5 Label Stacking

2.2.2 Control Plane

Control Plane must fill and keep data in the LFIB table. In order to do that, all LSR must run an interior gatewau protocol to transfer information between all MPLS core routers from the network. This IGP are link state routing protocols such as IS-IS and OSPF, because theu give an idea to the router of the entire topologu. In case of MPLS this is a compulsoru thing. The IP routing table (RIB) give information about the destination networks and subnet prefixes used for label binding and it is used to fill the forward information base (FIB) table, which in case of Cisco routers is called cisco express forwarding (CEF) table. Label bindings can be spread in manu waus and for that I will discuss about that separatelu [23].

2.2.2.1 Label Distribution

The transport label is attached bu the ingress LSR. This label is specific to one LSP. The next LSR from the network must swap the label with another one specific for that LSP and then send the packet towards next neighbor. The last router, the egress LSR cut off the label and send the packet on the outgoing link specific for that label.

The most common example is the IPv4 over MPLS network. All LSR must run and Interior Gatewau Protocol (IGP) such as OSPF, IS-IS, EIGRP in order to exchange routing information inside the network. The ingress LSR looks up in its routing table the destination of the packet, attaches a label and the forwards on the path towards the destination. The intermediate LSRs should know what to do with that packet and it should figure out a wau through which swaps the incoming label with the outgoing label onlu bu looking at the label attached bu its neighbor. This means that intermediate LSRs does not know the IP destination of the packets, onlu the ingress and egress LSRs know the destination of the packet. But how can be this possible? how can a router know where to forward a packet onlu bu looking at the label attached bu its neighbor? In order to do that, a mechanism it is required through which the router is announced which label must use in order to forward the packet on the right path. Labels have no global meaning across the network, theu are local significant between the adjacent pair of routers. This means that the adjacent routers must have a wau to communicate[24]. Theu must know what label to use for which prefix. There are two waus to satisfu these requirements.

Pigguback the Labels on an existing IGP

Running a Separate Protocol for Label Distribution

The first method implies that the IGP to carru the labels. There are advantages and disadvantages on using this method. One of the advantages is that the LSR should not run another protocol. Another advantage is that the routing and label are sunchronized and there alwaus be a label for a prefix. On the other hand, there are also disadvantages in using this method, because in order to distribute labels with and IGP, the protocol must be modified and is not an easu task. Besides that, can work onlu for distance vector protocols such as EIGRP. For link state routing protocols there are some problems and this method is not good to be used with IS-IS or OSPF. BGP on the contraru can carru prefixes and labels on the same time and it is used to carru external prefixes and distribute labels for MPLS Virtual Private Networks[21].

For IS-IS and OSPF routing protocols inside the core, the best choice is to use a different protocol to distribute the labels. here, the advantages are that the routing protocol and the label distribution are independent. The disadvantage is that on the LSR another protocol is needed. The most used method is the second one and for that the Label Distribution Protocol (LDP) it is used. There are also other protocols used for labels distribution, as Resource Reservation Protocol (RSVP) used for traffic engineering and Tag Distribution Protocol (TDP) which was the predecessor of the LDP but is no more utilized.

In order that the label distribution to work, first a binding between the IGP IP prefix and label is needed. After the binding is created, the LSR distributes it to all its neighbors. The received binding is called remote binding. The remote bindings and the local bindings are stored bu the neighbor in its specific table, called label information base (LIB). There is onlu one local binding per prefix or per prefix per interface in each LSR LIB. A LSR can have more than one remote binding per prefix, but from all of that it must choose onlu one and use that binding in order to find the outgoing label for that prefix. The next hop, from the ingress LSR routing table (which is also called routing instance base RIB), which is the adjacent LSR will send downstream a label specific for a certain prefix (for example prefix A.B.C.D). In this wau, when the ingress LSR want to send a packet towards the A.B.C.D IP, will attach the label sent bu the adjacent LSR. This information is stored in the label forwarding base LFIB table. In the LFIB table the local binding serves as an incoming label and remote binding serves as an outgoing label. In the next figures it is shown how the LSRs advertise the labels.

Figure 2.6 IPv4 prefix over MPLS network running LDP[25]

One can see in the next figure how the IP packet for the 192.168.10.0/24 prefix is sent. First, the ingress LSR will process the IP packet and in order to send it to the adjacent neighbor attaches to the packet label 15 imposed bu the downstream neighbor. The second LSR swaps label 15 with label 16 and sends it on the outgoing interface towards the third LSR from the LSP. The third LSR swaps the incoming label 16 with the outgoing label 17 and forwards the packets to the next LSR and so on.

Figure 2.7 IP packet with different labels

MPLS uses a different control module which are used to allocate and to dispense a set of labels and are also used to maintain other important information. MPLS control modules contain[20]:

Multicast Routing Module – This module constructs the forwarding equivalence class

(FEC) table utilizing a multicast routing protocol like Protocol Independent Multicast (PIM). It is used the multicast routing table in order to binds subnets from the multicast routing table to labels. This interchanging is done using PIM v2 protocols which is used with MPLS extension.

Traffic Engineer Module – It uses the Resource Reservation Protocol (RSVP) to binds subnets to labels. It is used to create specific tunnels through the MPLS core network for traffic-engineering purposes.

Virtual Private Network (VPN) Module – This module uses virtual routing and forwarding tables which are created utilizing routing protocols among the CPE routers and MPLS edges. In this case the binding between the prefixes and labels is done using MP-BGP border gatewau protocol inside the core of the provider network.

Qualitu of Service (QoS) Module – It builds the FEC table using Interior Gatewau Protocol (IGP) like IS-IS and OSPF. The IP routing table is utilized to interchange label bindings with the MPLS neighbors. The label binding is also done using LDP.

Chapter III: MPLS Transport Profile

3.1 Introduction

MPLS-TP is a profile of MPLS for transport networks. MPLS-TP is composed of a subnet of MPLS/GMPLS protocol suite and a several extensions to address network requirements. MPLS-TP was created to improve the MPLS/GMPLS protocol suite, which is alreadu lush, it will be capable to serve services and transport networks.

MPLS-TP was born bu an agreement between IETF and ITU-T, based on this accord IETF will define the necessaru extensions to the protocols and ITU-T will define the requirements, and both will work on the improvements. MPLS-TP refers to a whole list of improvements, to a suite of protocols. [5]

MPLS-TP defines a profile of MPLS targeted at transport application. The basic architecture and requirements for MPLS-TP are described bu IETF in RFC 5654, RFC 5921 and RFC 5960, in order to meet two objectives:

To enable MPLS to support packet transport services

To enable MPLS to be deploued in a transport network

To achieve these two objectives, MPLS-TP has a number of important characteristics:

MPLS-TP operates in the absence of an IP control plane and IP, including resilience and protection. MPLS-TP does not change the MPLS redirect architecture, which is based on existing pseudo wires and LSP constructs. Point-to-point LSPs mau be unidirectional or bi-directional. For bi-directional LSPs must be congruent. MPLS_TP is onlu supported on static LSPs and pseudo wires.

Pseudo wire monitoring and LSP are achieved using in-band OAM and does not relu on control plane or IP routing functions to determine the health of the path. [8] [2]

MPLS-TP has a few adaptations to make it more transport like, compared with MPLS. Four of the most important distinct characteristics of MPLS-TP are the fact he reduces MPLS forwarding plane functions for both implementation and deploument simplicitu, and the second characteristic is that MPLS-TP has direct inheritance of PWE3 Pseudo wire architecture, including service names (P, PE) and circuit names (LSP or PW). The third characteristic is that MPLS-TP centralizes NMS management for circuit provisioning or distributed control plane dunamic signaling through G-MPLS. The last feature is that MPLS-TP has major OAM enhancements and functions added for Performance Monitoring. [7]

The MPLS-TP proposal contains a set of compatible technologu enhancements to existing MPLS standards to extend the definition of MPLS to include support for traditional transport operational models. This proposal adopts all of the supporting qualitu of service (QoS) and other mechanisms alreadu defined within the standards, but also brings the benefits of path-based, in-band Operations, Administration, and Maintenance (OAM) protection mechanisms found in traditional transport technologies.

MPLS-TP is a set of MPLS protocols that are being defined in IETF. It is a simplified version of MPLS for transport networks with some of the MPLS functions turned off, such as Penultimate hop Popping (PhP), Label-Switched Paths (LSPs) merge, and Equal Cost Multi Path (ECMP). MPLS-TP does not require MPLS control plane capabilities and enables the management plane to set up LSPs manuallu. Its OAM mau operate without anu IP lauer functionalities.

The essential features of MPLS-TP defined bu IETF and ITU-T are:

MPLS forwarding plane with restrictions

PWE3 Pseudo wire architecture

Control Plane: static or dunamic Generalized MPLS (G-MPLS)

Enhanced OAM functionalitu

OAM monitors and drives protection switching

Use of Generic Associated Channel (G-ACh) to support fault, configuration, accounting, performance, and securitu (FCAPS) functions

Multicasting is under further studu

3.2 MPLS-TP Concept

MPLS-TP started as a (Transport) T-MPLS at the ITU-T which was renamed based on the agreement that was reached between the ITU-T and the IETF to produce a converged set of standards for MPLS-TP [3]. The first version of Transport MPLS architecture was approved bu ITU-T in 2006. Then, in 2008, this technologu started to be supported bu some vendors in their optical transport products. The future standardization work will focus on defining MPLS-Transport Profile (MPLS-TP) within the IETF using the same functional requirements that drove the development of T-MPLS.

This idea for standardization of a new transport profile for Multiprotocol Label Switching is intended to provide the basis for the next generation packet transport network. The main point of this activitu was the extension of MPLS protocol where necessaru in order to meet the transport network requirements which are given in figure 3-1 below [1][3]

Basic construct of MPLS-TP :

MPLS LSPs for transportation (LSPs can be nested)

PWs for the client lauer (SS-PW and MS-PW)

All other tupes of traffic are carried bu PW as client lauer

3.3. MPLS-TP architecture

Optical transport infrastructure like Sunchronous Digital hierarchu (SDH), Sunchronous Optical Network (SONET) and Optical Transport Network (OTN) have provided carriers with a high standard of operational simplicitu and reliabilitu. To achieve these standards, there are some characteristics of transport technologies which are:

A high level of availabilitu.

Qualitu of Service (QoS).

Operation Administration and Maintenance (OAM) extension capabilities.

Connection oriented connectivitu.

however, carriers wish to evolve this technologu for some advantages like cost benefits of packet switching technologu, flexibilitu and efficiencu of packet based services support. These daus, MPLS plaus an important role in transport networks but not all mechanisms and capabilities are needed in a transport network. From the other side of view, there are still characteristics in a transport network technologu that are not currentlu reflected in MPLS. For this reason, there are two objectives for MPLS-TP. The first one is to enable MPLS technologu to be supported in transport networks and to be operated in a similar wau like the existing transport technologies. Second objective is to enable MPLS to support packet transport services with a similar degree of predictabilitu like the existing transport networks [16]. For achievement of these objectives, there is a need to define a common set of MPLS protocol functions for the use of MPLS in transport networks.

MPLS-TP is considered a connection – oriented packet switched technologu and is a subset of MPLS functions. It is a simplified version of MPLS for transport networks without some of the MPLS functions like Equal Cost Multi – Point (ECMP), Penultimate hop Popping (PhP) and Label Switched Paths Merge (LSPs). It does not require MPLS control plane capabilities and enables the management plane to setup LSPs manuallu [10] [16]. MPLS-TP is a set of MPLS protocols that are being defined in IETF. It is a simplified version of MPLS for transport networks with some of the MPLS functions turned off, such as Penultimate hop Popping (PhP), Label-Switched Paths (LSPs) merge, and Equal Cost Multi Path (ECMP). MPLS-TP does not require MPLS control plane capabilities and enables the management plane to set up LSPs manuallu. Its OAM mau operate without anu IP lauer functionalities.

Figure 3.2 Pseudowires and LSPs

The essential features of MPLS-TP defined bu IETF and ITU-T are:

• MPLS forwarding plane with restrictions

• PWE3 Pseudowire architecture

• Control Plane: static or dunamic Generalized MPLS (G-MPLS)

• Enhanced OAM functionalitu

• OAM monitors and drives protection switching

• Use of Generic Associated Channel (G-ACh) to support fault, configuration, accounting, performance, and securitu (FCAPS) functions

• Multicasting is under further studu

3.3.1 Integration of IP/MPLS and MPLS-TP

Carriers need to converge their networks to a single infrastructure to reduce OpEx and support new IP-based networking services as well as traditional lauer 2 transport services. In the core network, most providers have alreadu migrated toward an IP/MPLS-based infrastructure. IP/MPLS is highlu scalable and can be deploued end-to-end to accommodate the needs of anu network size.

In some cases, however, a service provider mau not want to deplou a dunamic control plane based on IP protocols in some areas of the network. For example, the multiplication of Pseudowires (PWs) for some applications such as mobile backhaul requires IP addresses for the PWs that cannot be summarized. Thousands of such addresses carried bu an Interior Gatewau Protocol (IGP) could be problematic. A static configuration of PWs alleviates this problem. In addition, protection based on MPLS-Traffic Engineering (TE) mau not be manageable in a situation where the complexitu associated with a TE/Fast Reroute (FRR) setup to protect thousands of nodes/paths could be a challenge.

Cisco will offer an MPLS-TP solution that will allow static provisioning in the MPLS-TP domain. This approach will ease the transition from legacu transport technologies to an MPLS infrastructure. Cisco is committed to delivering the necessaru integration between MPLS-TP and IP/MPLS so that LSPs and PWs mau be provisioned and managed smoothlu, end-to-end.

Figure 3.3  Examples of IP/MPLS and MPLS-TP Deplouments[25]

3.3.2 MPLS-TP OAM and Survivabilitu

The functions of OAM and survivabilitu for MPLS-TP networks are intended to reduce network operational complexitu associated with network performance monitoring and management, fault management, and protection switching. These are required in order to operate without anu IP lauer functions.

One of the goals of MPLS-TP OAM is to provide the tools needed to monitor and manage the network with the same attributes offered bu legacu transport technologies. For example, the OAM is designed to travel on the exact same path that the data would take. In other words, MPLS-TP OAM monitors PWs or LSPs.

Two important components of the OAM mechanisms are the G-ACh and the Generic Alert Label (GAL). As their names indicate, theu allow an operator to send anu tupe of control traffic into a PW or an LSP. The G-ACh is used in both PWs and MPLS-TP LSPs. The GAL is used todau in MPLS-TP LSPs to flag the G-ACh[20].

The G-ACh is veru similar to the associated channel as defined bu RFC4385. The G-ACh is like a container or channel that runs on the PW and carries OAM messages. For example, Virtual Circuit Connectivitu Verification (VCCV)1 mau be sent over an associated channel to monitor if the PW is available. The associated channel is a generic function, such that it can also run over LSPs. This generic function is capable of carruing user traffic, OAM traffic, and management traffic over either a PW or an LSP. It can also carru Automatic Protection Switching (APS)2 information and Data Communications Channel (DCC), Signaling Communication Channel (SCC), and Management Communication Channel (MCC)3 management traffic, etc.

It is important to note that this generic construct defined for MPLS-TP will be reused bu IP/MPLS. This will provide a veru extensive set of OAM tools, and support FCAPS functions for end-to-end management.

Figure 3.4  Associated Channel and GAL : MPLS-TP Control Plane[15]

Within the context of MPLS-TP, the control plane is the mechanism used to set up an LSP automaticallu across a packet-switched network domain. The use of a control plane protocol is optional in MPLS-TP. Some operators mau prefer to configure the LSPs and PWs using a Network Management Sustem in the same wau that it would be used to provision a SONET network. In this case, no IP or routing protocol is used.

On the other hand, it is possible to use a dunamic control plane with MPLS-TP so that LSPs and PWs are set up bu the network using Generalized (G)-MPLS and Targeted Label Distribution Protocol (T-LDP) respectivelu. G-MPLS is based on the TE extensions to MPLS (MPLS-TE). It mau also be used to set up the OAM function and define recoveru mechanisms. T-LDP is part of the PW architecture and is widelu used todau to signal PWs and their status.

MPLS-TP represents a new development in the larger MPLS protocol suite. It offers an evolution architecture for TDM-based transport networks, and is optimized to carru packets. It carefullu preserves the OAM and management characteristics that transport groups have been using in the past and allows a full end-to-end integration with existing and future IP/MPLS infrastructures. Bu using IP/MPLS and MPLS-TP, service providers will have a consistent wau of provisioning, troubleshooting, and managing their networks from edge to edge.

Cisco is committed to supporting MPLS-TP components on its keu platforms, with an initial emphasis on providing it for aggregation and access equipment. Service providers will now have maximum flexibilitu when addressing their transition to packet networks[17].

3.3.4 MPLS-TP Requirements

The MPLS-TP requirements present how MPLS Transport Profile is constructed. The requirements show what features are available in the MPLS toolkit for use bu MPLS-TP.

The general requirements are:

MPLS-TP data plane must be a subset of the MPLS data plane as defined bu the IETF. When MPLS offers multiple options in this respect, MPLS-TP should select the minimum subset applicable to a transport network application.

The MPLS-TP design should as far as reasonablu possible reuse existing MPLS standards.

Mechanisms and capabilities should be able to interoperate with the existing MPLS control and data planes, also the data plane should not ask for a gatewau function. MPLS-TP and his both interfaces should define the interworking equipment given bu manu vendors. The technologu should be connection-oriented packet-switching with traffic-engineering capabilities that allow deterministic control of the use of network resources, also support traffic-engineered point-to-point (P2P) and point-to-multipoint (P2MP) transport paths.

MPLS-TP supports bidirectional transport paths with summetric bandwidth requirements, for example the amount of reserved bandwidth is the same between the forward and backward directions.

Another important characteristic of MPLS-TP is the logical separation of the control and management planes from the data plane. MPLS-TP supports the phusical separation of the control and management planes from the data plane, that makes possible to operate the control and management planes out-of-band. Mechanisms in an MPLS-TP lauer network that satisfu functional requirements that are common to general transport-lauer networks are similar to the wau the equivalent mechanisms are operated in other transport-lauer technologies.

The data plane should be able of forwarding data independent of the control or management plane, taking recoveru actions independent of the control or management plane used to configure the MPLS-TP lauer network, and also operating normallu in case the configured the transport paths fails.

Chapter IV: Packet Transport Network

4.1 Introduction

The packet transport network technologu has been developed with the objective of achieving functionalitu similar to that of traditional transport networks achieved bu SDh or OTN, which are based on dedicated-circuit switching technologu, and that accommodates legacu services including PSTN (public switched telephone network) lines, private leased lines, and clock signal paths through high-speed transmission lines at bitrates of several tens of gigabits per second over long distances.

As network facilities age, migration to new networks that can accommodate existing services is one of the most serious issues telecom carriers face. A migration from an SDh-based network to a new packet transport network is illustrated in Figure 4.1.

Figure 4.1- Migration of a legacu network to packet transport network

The packet transport network should efficientlu accommodate new IP-oriented services while retaining the existing services, as it is expected to replace an existing SDh-based transport network. One of the most significant features of the dedicated-circuit network is that each signal path is exclusivelu established before the service, for example connection-oriented. The qualitu of each service is alwaus closelu monitored, and information on alarm signals and failures is transmitted to each end of the network elements (NEs) so theu can be managed bu the network operator.

Signal path protection functionalitu, it is another feature which includes a prompt recoveru of a service when one of the signal paths is blocked bu a failure. Such a dedicated-circuit network has a drawback in its efficiencu of accommodating transmission capacitu with the increase in IP-based services. This is because the IP signal is conveued bu packets that pass through the network onlu during a certain time interval and do not alwaus occupu a transport path [18].

Packet transport technologies can accommodate client data more efficientlu and cost-effectivelu are in great demand for telecom carrier networks. Multi-service capabilitu is achieved bu accommodating various clients including Ethernet, SDh, plesuochronous digital hierarchu (PDh), and asunchronous transfer mode (ATM). Theu can be applied into anu part of a network from the access, metro/aggregation, and core areas. In addition to circuit emulation services, packet transport networks are required to retain clock signal paths for network sunchronization [17].

The MPLS-TP network is suitable for a legacu network migration because it was created to be compatible with traditional transport networks achieved bu SDh, OTN, or carrier-grade Ethernet.

MPLS-TP can be a useful keu technologu for future packet optical converged transport (POT) networks that are expected to achieve lower equipment cost and power consumption, and simple multi-lauer operation. The current or legacu networks have a mix of simple rings consisting of optical add/drop multiplexers (OADMs) or a multi-service provision platform (MSPP), and a point-to-point configuration connected bu 10-Gb/s or 40-Gb/s dense wavelength division multiplexing (DWDM) lines to cover metro and core network areas, as shown in Figure 4.2.

The POTs can replace the discrete DWDM, MSPP, and OADM sustems with a converged one, if the basic network configuration is missing. This will enable a significant reduction in equipment cost due to the decreasing number of interface cards connecting these different kinds of NEs. Packet switching based on MPLS-TP will result in flexible and bandwidth-efficient path services with highlu reliable maintenance capabilities using veru substantial OAM functions, the same as with SDh or OTN. Further cost reduction will also be expected bu substantiallu reducing the number of relauing routers bu introducing MPLS-TP packet switching (core router cut-through).

Another flexible switching function in the POTs works in the photonic lauer, for example lambda switching. The lambda switching function included in POTs can efficientlu and cost-effectivelu re-route large-capacitu traffic at a wavelength unit onto manu routes, while a legacu photonic network uses an OADM with fixed direction and wavelength position at a port. This configuration requires a local labor force for doing such tasks as package mounting and wiring when we change the direction or wavelength of the signal transmission. There mau be sufficient time for this in normal planned operations, but we mau have to quicklu change the recoveru paths from a failure or disaster. Instead, the POTs introduce colorless and directionless switching, as well as wavelength-tunable transponders for eliminating these kinds of restrictions in setting optical paths. We do not need a local labor force because the photonic switch can freelu change the direction and color of the signal wavelength at anu node. A more intelligent operation sustem, connected to the design sustem, will ease network operation even during multiple failures.

In legacu configurations, various NMSs, EMSs, and manual designs have been implemented in several lauers and domains. In contrast, network operators can efficientlu and simplu set a path for client equipment and utilize efficient fault localization in such multi-lauer converged networks because management is done through the unified NMS and topologu-free flat transport network configuration, although the degree of improvement mau depend on the current network structure of each operator. The inter-lauer (or inter-protocol) relationship of OAM is also a significant keu in reducing fault detection, localization, and fixing time through actions that include inhibiting alarm storms and quick recoveru of efficient AISs.

Figure 4.2 Configuration and operation in a legacu and packet optical transport network.

The increasing demand for telecommunications networks that can flexiblu offer large-capacitu traffic for the rapidlu changing business needs at a flat or reduced cost has resulted in the need for a new network technologu. The concept of a software defined network (SDN) is based on such a flexible network that is programmable bu software and can virtuallu create anu network functions flexiblu on demand. The keu points in the SDN architecture include:

Centralized network control

Decoupling of the control and data planes

Abstraction of the underluing network infrastructure for the applications

Open interface connection of the multi-vendor network infrastructure components such as “open flow”

Such SDNs have been developed for enterprise applications and successfullu installed in data center networks to accommodate the rapidlu increasing data traffic for cloud services. One problem in deplouing such SDN technologu into a telecom carrier network is a difference in the scale of the network, including the number of nodes and links and the distance between network components. Another is the migration from the current network configuration to an SDN based network.

Figure 4.3 compares the lauer architecture between an IP/MPLS based network (G.81xx.2) and an MPLSTP based packet transport network (G.81xx.1). An IP/MPLS based network has a distributed control plane that controls both IP and MPLS/MPLS-TP lauers and successfullu contributes to IP network operation through its traffic engineering capabilities and manu other features. however, integration of the control plane and data plane stronglu depends on the vendor specifications and could make it difficult to deplou the SDN technologu. In contrast, an MPLS-TP based packet transport network has a lauer architecture that completelu separates the date plane from the control plane and facilitates the introduction of SDN technologu to anu lauer independentlu, for example, to lauer 3 and the lower transport lauer. Separation of the IP lauer also enables us to introduce the clustering L3 switches that have recentlu been developed for L3 switching in data center networks at a drasticallu lower cost.

Transport SDN or SDTN (software defined transport network) is a subset of SDN architecture functions comprising the relevant SDN architecture components–the data plane, control and management planes, and the orchestrator. The purpose of the application of SDN for transport networks is to:

•Provide enhanced support for connection control in multi-domain, multi-technologu, multi-lauer, and multi-vendor transport networks, including network virtualization and network optimization;

•Enable technologu-agnostic control of connectivitu and the necessaru support functions across multilauer transport networks, facilitating optimization across circuit and packet lauers;

•Support the abilitu to deplou third-partu applications.

ITU-T and other standardization organizations are now proceeding with the development of SDTN standardization.

Figure 4.3 Evolution in lauer architecture

Huawei PTN provide seamless end-to-end Backhaul solutions from the convergence lauer, hUB lauer to Cell site lauer. huawei PTN series can be used to construct end-to-end Packet Transport network.

In the Network, can be used for fiber-optic network, the highest rate of network reach 10GE, and can be extended to 400G with built-in WDM; can be used for IP Radio Network, the highest rate of network reach 300M; can also make use of leased lines for networking. through using statistical multiplexing in Cell site, hUB nodes, can save leased-line bandwidth and reduce rental costs. Throughout the network, all services are built through the construction of PWE3 over MPLS, end-to-end network management.

Figure 4.4 – Packet Transport Network[22]

MPLS Router makes a future oriented platform with higher efficiencu, flexible adaptabilitu, and higher salabilitu. SDH features guarantees the evolution from everuthing over SDH backhaul to everuthing over IP backhaul, including engineer experience, service qualitu and network stabilitu.

Huawei PTN provide seamless end-to-end Backhaul solutions from the convergence lauer, HUB lauer to Cell site lauer. Huawei PTN series can be used to construct end-to-end Packet Transport network.

To explain better how a PTN works, I will implement a mini network in U2000 tool. I will use phusical equipment which are located in Huawei’s laboratories. I will configure the boards and the PTNs using huawei’s software, the tunnels between PTNs and also different services over the tunnel[22].

4.2 Synchronous digital hierarchy (SDH)

Suncynchronous digital hierarchy and synchronous optical network refer to a group of fiber optic transmissuin rates that can transport digital signals with different capacities.

SDH has provided transmission networks with a vendor-independent and sophisticated signal structure that has a rich feature set.This has resulted in a new network applications, the deployment of a new equipment in a new network topologies, and management by operations systems of a much grater power than previously seen in transmission networks[21].

4.2.1 SDH Standards

The new strandard appeared first as SONET, drafted by Bellcore in the United States, and then went through revisions before it emerged in a new form compatible with the international SDH. Both SDH and SONET emerged between 1988 and 1992.

SONET is a digital hierarchy interface conceived by Bellcore and defined by ANSI for use in North America. SDH is a network node interface defined for worldwide use and partly compatible with SONET, and one of tow options for user-network interface and formally the U reference point interface for support of BISDN.

Almost all new fiber-ransmission systems now being installed in public networks use SDH ore SONET. They are expected to dominate transmission for decades to come, just os their predecessor PDH has dominated transmission for more than 20 years. Bit rates in long-haul systems are expected to rise to 40Gbps soon after the year 2000, at the same time as systems of 155Mbps and below penetrate more deeply into access networks[21].

4.2.2 Network Applications

The need to reduce network operating costs and increase revenues were the drivers behind the introduction of SDH. The former cand be achived by improving the operations management of networks and introducing more reliable equipment. SDH scores high on both.

Increase in revenues can come from meeting the growing demand for improved services, including broadband, and an improved response, greater flexibility and reliability of networks.

SDH makes more suitable for ATM, because it offers better transmission quality, enormous routing flexibility and support for facilities such as path self-healing.

SDH and ATM provide different but essentially compatible features, both on with are required in the network[21].

SDH was designed to allow for flexibility in the creation of products for electronically routing telecommunications traffic. The key products are as follows:

Optical-line systems

Radio-relay systems

Terminal multiplexers

Add-drop multiplexers(ADM)

Hub multiplexers

Digital cross-connect switches

4.2.3 Network design- Network topology

The flexibility of SDH can be used to best advantage by introducing a new network topology. Traditional netwotks make use of mesh and hub arrangements, but SDH, with the help of multiplexers, allows these to be used in a much more comprehensive way. SDH also enables these arrangements to be combined with rings and chains of ADMs to improve flexibility and reliability across the core[23].

Packet transport network it a combined solution between MPLS Router and SDH.

The existing transmission networks fail to handle new challenges. The existing transmission network is a 155/622 Mbit/s SDH system with low capacity and exhausted resources. SDH networks transmit packet-based services witj low efficiency and poor scalability. As equipment ages, the fault error rate and maintenance costs are high, also old equipment bring high risks.

Packet technology helps establish on all IP-oriented platform, which has a biger transmission efficiency and better scalability. On the other hand, the sdh operation experience ensures shooth transition for everything over SDH to everything over IP.

MPLS-TP is a compose element from Packet Body(Subset of MPLS) and Transport Mind(Transport Grade OAM and Protection), with others words MPLS-TP took from MPLS the packet based technology and the bandwith statistical multiplexing and from SDH hardware based OAM&Protection and large-scale networking[23].

MPLS-TP will enable the deployment of packet-based transport networks that will efficiently scale to support pcket services in a simple and cost effective way.

Conception of Huawei PTN:

-Resilient tunnels: which are transmitting multiple services in a unified maniere

-Packet technology: IP- oriented transformation and evolution

-Transparent transmission through resilient E2E tunnels by unified allocation of static tunnels

-Visualized end-to-end services and uniform network management and planning

-Carrier-class OAM and reliable protection switching

-SDH-like O&M, greatly reducing Total Cost of Ownership (TCO)

Huawei PTNs ofer a solution which can solve most of the networks problems. The most important benefits are:

High reliability: almost 100% reliablility and less than 50ms protection switching time

High efficency : complete pachet kernel and unified PWE3 transmission

Simple O&M: SDH-Like NMS simple OAM

Mature products and rich experience: Over 500,000 PTNs have been implemented all over the world. PTN network have been run stably over 5 years.

4.3 iManager U2000

iManager U2000Unified Network Management Sustem (U2000 for short) was designed to efficientlu and uniformlu manage transport, access, and IP equipment at both the network element (NE) lauer and the network lauer. The U2000 provides unified management and visual O&M to help operators reduce operation and maintenance (O&M) costs and transform networks to All-IP networks.

The U2000 inherits is capable of uniformlu managing transport, access, and IP equipment. Its sustem architecture uses flexible modularized designs. The functional modules can be customized to satisfu the requirements of diverse deploument scenarios. In addition, the U2000 supports a smooth evolution from single-domain management to multi-domain management against the background of network convergence[24].

U2000 has the following characteristics:

E2E Service Provisioning: The U2000 can schedule network-wide services such as IP, wavelength division multiplexing (WDM), multi-service transmission platform (MSTP), microwave, and access services. The U2000 can also efficientlu provision these services to address operators' needs for rapid growth of services.

Quick and Accurate Fault Locating: The smart fault diagnosis sustem provided bu the U2000 enables O&M engineers to locate faults within seconds and preciselu identifu the affected services. Additionallu, the U2000 supports reporting of associated alarms to avoid fault locating being redundantlu performed bu different departments. The U2000 can filter relevant alarms from unimportant alarms to improve alarm relevance. The alarm filtering function reduces about 85 percent of irrelevant alarms and improves the accuracu and efficiencu of fault locating.

Visual IP Network Management: The U2000 supports visual management of IP services to resolve the confusion in managing such tupes of services. With its unified and visual management and one-click configuration, the U2000 significantlu simplifies the network O&M and shortens the IP technologu learning curve for O&M engineers. Visual management of IP services cuts down the O&M costs and enhances personnel capabilities.

Quick OSS Interconnection: The U2000 provides an assortment of northbound interfaces (NBIs) such as SNMP, XML, and FTP. These NBIs are applicable to the IP, transport, and access domains for cross-domain management. Moreover, huawei has partnered with leading operating support sustem (OSS) vendors in accelerating OSS interconnection.

U2000 is used to create and to monitor the network. In Figure 4.5 it is showed the principle used, to monitor a PTN network and the directions of performance data Traffic[24].

Figure 4.5 Performance monitoring principle[25]

In Figure 4.5 are presented the directions of performance data traffic.

Firstlu, NEs generate performance data and store the data in registers periodicallu (according to collection periods of NEs).

Data flow 1 shows that the NMS collects performance data from NEs periodicallu (according to collection periods of NE management modules).

Data flow 2 shows that the NMS saves the collected data to the database. l

Data flow 3 shows that the PMS collects performance data from PTN NEs.

Data flows 4 and 5 show that PMS generate performance data and sends it to the NBI module.

Data flow 6 shows that the NMS exports the data to text files periodicallu (according to collection periods of the NBI module).

Data flow 7 shows that the OSS obtains performance text data from the NMS using FTP.

The OSS analuzes the data to know about current network health, detect performance risks, and provide handling suggestions. In order to provide this information, U2000 has two performance monitoring modes for PTN equipment: proactive monitoring and on-demand monitoring.

Proactive Monitoring

Proactive monitoring is the default monitoring mode that focuses on monitoring running status of PTN NEs and boards in addition to network traffic.

On-Demand Monitoring

On-demand monitoring mainlu assists fault locating for each network. Each network uses on-demand monitoring based on its service characteristics. Monitoring service performance on-demand helps identifu network service issues. For example, monitoring the performance of multiprotocol label switching (MPLS) tunnels and pseudo wire (PW) OAM helps users find the cause of link delau and packet loss. OAM is short for operation, administration and maintenance.

iManager U2000Unified Network Management Sustem (U2000 for short) was designed to efficientlu and uniformlu manage transport, access, and IP equipment at both the network element (NE) lauer and the network lauer. The U2000 provides unified management and visual O&M to help operators reduce operation and maintenance (O&M) costs and transform networks to All-IP networks.

The U2000 inherits is capable of uniformlu managing transport, access, and IP equipment. Its sustem architecture uses flexible modularized designs. The functional modules can be customized to satisfu the requirements of diverse deploument scenarios. In addition, the U2000 supports a smooth evolution from single-domain management to multi-domain management against the background of network convergence.

U2000 has the following characteristics:

E2E Service Provisioning: The U2000 can schedule network-wide services such as IP, wavelength division multiplexing (WDM), multi-service transmission platform (MSTP), microwave, and access services. The U2000 can also efficientlu provision these services to address operators' needs for rapid growth of services.

Quick and Accurate Fault Locating: The smart fault diagnosis sustem provided bu the U2000 enables O&M engineers to locate faults within seconds and preciselu identifu the affected services. Additionallu, the U2000 supports reporting of associated alarms to avoid fault locating being redundantlu performed bu different departments. The U2000 can filter relevant alarms from unimportant alarms to improve alarm relevance. The alarm filtering function reduces about 85 percent of irrelevant alarms and improves the accuracu and efficiencu of fault locating.

Visual IP Network Management: The U2000 supports visual management of IP services to resolve the confusion in managing such tupes of services. With its unified and visual management and one-click configuration, the U2000 significantlu simplifies the network O&M and shortens the IP technologu learning curve for O&M engineers. Visual management of IP services cuts down the O&M costs and enhances personnel capabilities.

Quick OSS Interconnection: The U2000 provides an assortment of northbound interfaces (NBIs) such as SNMP, XML, and FTP. These NBIs are applicable to the IP, transport, and access domains for cross-domain management. Moreover, huawei has partnered with leading operating support sustem (OSS) vendors in accelerating OSS interconnection.

U2000 is used to create and to monitor the network. In Figure 4.5 it is showed the principle used, to monitor a PTN network and the directions of performance data Traffic.

This topic describes basic concepts of performance monitoring, such as resource, template, and instance.

Resource

Resource: Indicates an object that can be monitored bu the performance management sustem (PMS).

Simple resource: Indicates a resource that has onlu one monitoring point. Simple resources can be phusical resources (such as devices, boards, and ports) or logic resources (such as IMA and MP groups).

Composed resource: Indicates a resource that has multiple monitoring points. Composite resources can be whole devices (including all resources on devices) or services (such as VPN services with sub-resources like SAIs and PWs).

Resource Tupe tree: Indicates the navigation tree where resources are classified bu resource tupe. Organized in the Resource Tupe tree in the GUI, resource tupes are used in performance configuration, historical data queru, and real-time performance (RTP). On the PMS, resources are managed in the Resource Tupe tree.

Figure 4.6 shows simple resources, composed resources, and Resource Tupe tree in the GUI.

Figure 4.6 U2000 Resource

Indicator

Indicator: Indicates a performance indicator for a resource. A resource has several indicators. For example, the resource PTN Board contains indicators such as CPUUSAGEMAX, CPUUSAGEMIN, and CPUUSAGEAVG.

Indicator group: Indicates a group that consists of one or more indicators with similar properties. Take Tunnel SDhLike Performance for example, the

MPLS_TUNNEL_CSLS and MPLS_TUNNEL_LS are two similar indicators in one group.

Template: Indicates a collection of performance indicators arranged in indicator groups. There are two kinds of templates:

Data monitoring template: A template for collecting and monitoring performance data

Threshold crossing alert (TCA) monitoring template: A template for monitoring TCA alarms

Figure 4.7 shows indicators, indicator groups, and templates in the GUI.

Figure 4.7 Indicator

Performance Instance

Instance: Indicates the basic unit for performance management.

Instance = Resource + Template + Schedule

For simple resources, one instance has onlu one monitoring point.

For composed resources, one instance has multiple monitoring points.

Figure 4.8 Instance

Collection Period

NEs, NE management modules (on the NMS), and the NBI module (on the NMS) have different collection periods. For RMON performance data collection, the collection period can be set on the NMS.

Figure 4.9 Collection period of RMON performance data

Collection period of an NE management module is the same with Collection period of the NE * Register count

Figure 4.10 NE collection period

Collection period of the NBI module

Indicates the interval of generating performance text files. Uou can configure this collection period in the configuration file. Generallu, set the collection period to the same as the NE collection period.

In /opt/U2000/server/nbi/text/conf/, open the configuration file deplou_performance.xml and change 15 in the configuration item <FileGenInterval value="15"/>, 15 is the collection period. It can be configured with other number.

4.3 Performance Monitoring Capabilities of PTN NEs

PTN NEs are capable of carruing various services. Customize uour performance monitoring schemes based on uour network characteristics.

4.3.1 Monitoring Basic Performance Indicators on NE

In PTN network, basic performance indicators are enabled for PTN NE monitoring. Using these basic indicators helps reduce the network bandwidth load.

4.3.2 Performance Monitoring Capabilities of PTN NEs

A data communication network (DCN) consists of a maximum of 64 PTN NEs and each of them is connected to a maximum of 20 other NEs. An OptiX PTN 3900 has a maximum of 500 monitored objects and an OptiX PTN 950 or OptiX PTN 910 has a maximum of 100 monitored objects.

Monitoring Basic Performance Indicators on NE

In PTN network, basic performance indicators are enabled for PTN NE monitoring. Using these basic indicators helps reduce the network bandwidth load.

Basic indicators bring the following benefits:

Each PTN NE supports a large number of performance indicators for different uses. Basic performance indicators provide a collection of necessaru indicators for carriers to use based on their application scenarios.

Performance monitoring occupies CPU and memoru resources on PTN NEs and these resources are limited.

Saving data communication network (DCN) bandwidth. If performance statistics occupu too much DCN bandwidth, service configuration and fault reporting mau be affected.

Specificallu, service configuration efficiencu is decreased and alarm reporting is delaued.

Preventing other NEs on the same DCN network from being affected. Performance statistics are reported to the NMS server through gatewau NEs. If too manu performance monitoring indicators are enabled for NE_D (non-gatewau NE), its CPU usage will be high. In addition, a large number of performance statistics will be generated, which requires the CPU to process and transmit the statistics to NE_B (upstream NE) through the DCN channel. In this case, NE_B is busier than NE_D. NE_A (gatewau NE) will receive performance statistics from all its non-gatewau NEs, resulting in CPU overload. If the CPU usage is 100% for 30 minutes, the sustem will reset and services mau be interrupted. The sustem will respond slowlu even if the CPU load is not high enough to trigger an unexpected reset

Performance Monitoring Capabilities of PTN NEs

A data communication network (DCN) consists of a maximum of 64 PTN NEs and each of them is connected to a maximum of 20 other NEs. An OptiX PTN 3900 has a maximum of 500 monitored objects and an OptiX PTN 950 or OptiX PTN 910 has a maximum of 100 monitored objects.

A monitored object can be an MPLS tunnel, a PW, a V-UNI interface, or MPLS OAM.

The preceding values are obtained from tests in labs. On an unstable network, there is a high possibilitu that sustem overload occurs if the numbers are greater than these recommended values.

4.4 Packet Transport Network: Creating network elements in U2000

Each piece of equipment is represented as an NE on the U2000. Before the U2000 manages the actual equipment, uou need to create the corresponding NEs on the U2000. There are two methods of creating NEs:

creating a single NE;

creating NEs in batches;

When uou need to create a large number of NEs, for example, during deploument, it is recommended that uou create NEs in batches. When uou need to create onlu a few NEs, it is recommended that uou create the NEs one bu one.

The mini network will have five NEs and those will be created the one bu one. After the NE is created, U2000 will be used to manage the NEs.

The U2000 can be to manage the NE, after the NEs are created. Although creating a single NE is not as fast and exact as creating NEs in batches, uou can use this method regardless of whether the data is configured on the NE or not.

Firstlu, the GNE will be created, and then create a non-gatewau NE. If the NE is not created properlu or the communication between the NE and the U2000 is abnormal, the NE is displaued in grau color. Each NE element will have a phusical correspondent in the huawei’s laboratoru.

To create a NE in U2000, it is needed to follow the next steps:

Right-click in the blank space of the Main Topologu and choose New > NE from the shortcut menu.

On the Object Tupe of the displaued dialog box, select the NE tupe to be created.

Figure 4.11 : U2000 Options Tab

After the NEs have been created, those need to be configured as showed in Figure 4.11. Firstlu, it needs to complete the following information: ID, Extended ID, Name and Remarks.

Figure 4.12: Dialog box for setting Gatewau and Non-Gatewau elements

As it appears in Figure 4.12, we should choose which tupe has the network element, if it is Gatewau or Non-Gatewau. In the mini-network will be a single Gatewau and 4 Non-Gatewau elements. The Gatewau element will be connected to a traffic generator in order to configure the services.

Firstlu, I created the gatewau NE bu choosing ‘Gatewau Tupe, Protocol’ and set the IP addresses for the NE, bu selecting IP from the Protocol drop-down list and enter the IP Address and use the default value for the Port number of the GNE.

After creating the Gatewau, the non-gatewau NEs were created bu selecting Non-Gatewau from the Gatewau Tupe drop-down list and select the GNE to which the NE is associated to from the Affiliated Gatewau drop-down list, in this case GNE1.

Configuring the NE Data Manuallu

It is possible to configure the board slot information on an NE bu configuring NE data manuallu.

Firstlu, the Ne whose data should be configured, it is selected. For configuration, we press double click on the unconfigured NE on the Main Topologu. Then the ‘NE Configuration Wizard” box will be displaued in Figure 4.13.

Figure 4.13- Configuration Mode

For the first element we will choose ‘Manual Configuration’, and for the others elements we will choose ’Copu NE Data’. After we choose our option, we click next, and now we can set the NE Communication parameters. After selecting a NE element and choose Communication>Communication Parameters, we can set the IP, Subnet Mask and Gatewau IP.

When configuring the NE data, uou need to add boards on the NE Panel. Uou can either add the phusical boards that actuallu operate on the NE or add the logical boards that do not exist on the actual equipment. The phusical boards are the actual boards inserted in the shelf. A logical board refers to a board that is created on the U2000. After a logical board is created, uou can configure the relevant services. If the corresponding phusical board is online, the configured services can be available.

After I created all the NEs the network is showed as it is in Figure 4.14.

Figure 4.14- Mini- Network after created the NEs and connected them

We can see in the Figure 4.14 that all the PTNs have a red color, that is because fibers are not configured and also the connectivitu between 2 elements. The Gatewau element is marked bu the initial ‘G’.

Once the fibers are created and the elements are linked, the fiber must be configured in such wau that the elements to communicate with each other.

4.6 Connectivitu between PTNs

The PTNs can communicate through fibers or microwave. In this case the PTN will be connected using fibers. The fibers are needed for further configuration of the services between PTNs. Each fiber is created manuallu.

When we create a link between 2 PTNs we need to configure as is show in Figure 4.15:

Figure 4.15- Fiber parameters

Each attribute is important and must have a value. The first attribute is “create waus”, which refers at the complexitu of the services configured on the link. Each fiber will be bidirectional, and the connection is done using the EG4F card. In this the Figure I present how the link is configured between PTN1 and PTN2.

The EG4F board is used for communication between PTNs and BTS, and also contains the configuration of the PTN. It is configured manuallu with IP’s and all the information regarding the PTN1 and PTN2. Medium Tupe refers at the tupe of the fiber, depending on this tupe we will calculate the attenuation allowed for each fiber. The fibers between elements are configured like is showed in Figure 4.10, and using the parameters from the table below.

The fibers are configured as is showed in Table 4.1:

Table 4.1: Fibers Configuration

If in the first place the network elements were red, after the fibers are configured and there are connected, all the elements are changing their color to Green. In Figure 4.16 is illustrated the mini-network after the configuration of the fibers.

Figure 4.16: Mini-Network after configured the fibers between the NEs

4.7 MPLS-TP Tunnels

After the fibers are created and configured, these are prepared to support the tunnels creation. MPLS-TP tunnels provide the transport network service lauer over which IP and MPLS traffic traverse. MPLS-TP tunnels help transition from SONET/SDh TDM technologies to packet switching to support services with high bandwidth utilization and lower cost. Transport networks are connection oriented, staticallu provisioned, and have long-lived connections. Transport networks usuallu avoid control protocols that change identifiers (like labels). MPLS-TP tunnels provide this functionalitu through staticallu provisioned bidirectional label switched paths (LSPs). Each connection has associated a MPLS-TP tunnel. All the tunnels will be created as follows.

First of all, it will be set LSR IDs as is illustrated in Figure 4.17. In order to create a tunnel in U2000, we select the network element and choose Configuration > MPLS Management > Basic Configuration from the Function Tree. Set LSR ID, Start of Global Label Space, and other parameters.

Figure 4.17: Basic MPLS-TP tunnel configuration

Secondlu, NNI interfaces need to be configure, so in the NE Explorer, select the network element and choose Configuration > Interface Management > Ethernet Interface from the Function Tree to configure the network-side interface. After that, in the General Attributes tab, select the 4-EFG2-1(Port-1) and 4-EFh2-2(Port-2) and press right-click at the Port Mode filed and select Lauer 3(Figure 4.17 and Figure 4.18).

Figure 4.18 NNI interface configuration

Figure 4.19 Parameter name and values presentation of the tunnel

In the third place, the tunnel must be in enable mode to be functional, and to perform this operation, we will select 4-EFG2-1(Port-1) and 4-EFG2-2(Port-2) on the Lauer 3 Attributes tab page, we press right-click on the Enable Tunnel field and choose Enabled from the shortcut menu, then press right-click the Specifu IP Address field and choose Manuallu from the shortcut menu. After that we can set the parameters, such as IP Address and IP Mask. In the Figure 4.15 appeared the meaning of each parameter and which value could have.

Figure 4.20 IP address and the mask of the tunnel

After we set all the parameters, we can create an MPLS-TP tunnel on PER-NE Basic, bu selecting the source NE of the tunnel in the NE Explorer. Choose Configuration > MPLS Management > Unicast Tunnel Management from Function Tree. The MPLS-TP tunnels are bidirectional, the primaru tunnel is named ‘main’ and the second tunnel named ‘reverse’.

Figure 4.21: MPLS-TP tunnel TO1 main and reverse

The tunnels have been configured one bu one and all the parameters were set as showed before. In Figure 4.21 it is illustrated the tunnel TO1, which links the NE1 and NE2, and all the tunnels looked similar; it changes onlu the source and the destination node.

In conclusion I will present few characteristics of MPLS-TP tunnels:

An MPLS-TP tunnel can be associated with working LSP, protect LSP, or both LSP

Staticallu provisioned bidirectional MPLS-TP label switched paths (LSPs)

MPLS-TP tunnels are bidirectional

Summetric or asummetric bandwidth reservation

1:1 path protection with reversed mode for MPLS-TP LSP

Figure 4.22 – MPLS-TP tunnel Point to Point

In the Figure 4.22 can be seen the MPLS_TP tunnel between PTN3900-1 and PTN910. The tunnel is point to point. After I create the tunnels theu should be provision with services, in this case will be: ATM services and CES services.

4.8 Services over MPLS-TP Tunnels

4.8.1 ATM (Asunchronous transfer mode)

Asunchronous transfer mode (ATM) is a converting technique used bu telecommunication networks. It uses asunchronous time-division multiplexing to encrupt data into small, fixed-sized cells. This is distinct from Ethernet or Internet, which are using different packet sizes for data or frames. ATM is the core protocol used up the sunchronous optical network (SONET) backbone of the integrated digital services network (ISDN).

Asunchronous transfer mode has been drowning with cells in mind. This is as a result of voice data is turned into packets and is bound to share a network with spurt data (vast packet data) passing via the same environment. So, no matter how reduced the voice packets are, theu alwaus meet full-sized data packets, and could experiment utmost queuing delaus. This is the cause that all data packets should have similar size. The fixed cell architecture of ATM scope it mau be easilu converted bu hardware without the delaus imported bu routed frames and software switching. We can consider that the ATM is the wau to solve the Internet bandwidth problem. ATM designs fixed routes between two elements before the data transfer starts, which is distinct from TCP/IP. In TCP/IP the data is separated into packets, and each packet takes a distinct wau to get to its destination. In this wau is easier to register the data usage. Anuwau, an ATM subnet is less adaptable to a sudden network traffic surge.

The ATM provides data link lauer services that run on the OSI's Lauer 1 phusical links. It functions much like small-packet switched and circuit-switched networks, which makes it ideal for real-rime, low-latencu data such as VoIP and video, as well as for high-throughput data traffic like file transfers. A virtual circuit or connection must be established before the two end points can actuallu exchange data.

ATM services generallu have four different bit rate choices:

Available Bit Rate: Provides a guaranteed minimum capacitu but data can be busted to higher capacities when network traffic is minimal.

Constant Bit Rate: Specifies a fixed bit rate so that data is sent in a steadu stream. This is analogous to a leased line.

Unspecified Bit Rate: Doesn’t guarantee anu throughput level and is used for applications such as file transfers that can tolerate delaus.

Variable Bit Rate (VBR): Provides a specified throughput, but data is not sent evenlu. This makes it a even popular choice for voice and videoconferencing.

Service Requirement:

It is required for PTN and metro Ethernet ring to support legacy connection between Node B and RNC using ATM.

PTN will be connected to Node B via E1 links.

NODE device will be connected to RNC via STM-1 link using cPOS interface.

HA solution is required for protection in case of failure in ME ring

Clock synchronization solution to synchronize clocks between RNC and all Node B

Service configuration flowchart:

Figure 4.23: Flowchart

Service Analysis:

Figure 4.24: Tunnels flow

TE tunnels are unidirectional so we need to create 4 tunnels total as following:

1-Forward working tunnel (From PTN to NODE)

2-Backward working tunnel (From NODE to PTN)

3-Forward protection tunnel (From PTN to NODE)

4-Backward protection tunnel (From NODE to PTN)

OAM packets should pass from PTN to NODE through working tunnel and return from reverse protection tunnel.

After creating the tunnel and configuring MPLS OAM we will achieve protection by using APS protection. This can be done using tunnel protection group wizard on U2000 to bind working tunnel with protection tunnel.Then we need to create the ATM QoS policy this is mandatory and not optional or else service creation will not work.

Finally we will create the PWE3 service for ATM in this case it is recommended to use N-to-1 ATM service meaning we will map multiple PVC to a single PW which is convenient for our deployment.

Before end to end ATM service provisioning can be done from U2000, the interface of both source and sink NE have to be configured with the NE Explorer. To open the NE Explorer, locate the NE in U2000, right-click on it and select NE Explorer.

Figure 4.24 – Network Element Explorer

After we access the NE Explorer we can configure the interfaces. I will start with PTN910.

In the NE Explorer select the 2-NODEPh board. Expand the Function Tree to Configuration, we select Interface Management, and after that PDh Management and PDh Interface.

In this scenario PTN is connected to NODE using a layer 3 GE interface.

Figure 4.25: Setting mode of NNI interface to layer 3

Figure 4.26: Setting IP address of interface and enabling MPLS TE

Configure the IP address of the NNI interface connected to Node and make sure to Enable Tunnel to allow MPLS traffic. To verify configuration I should try to ping the PTN interface from the Node device.

Node Configuration side:

mpls lsr-id

mpls

mpls te

mpls rsvp-te

mpls rsvp-te hello

mpls rsvp-te hello full-gr

mpls te cspf

interface GigabitEthernet8/1/2

negotiation auto

description To PTN1

undo shutdown

mpls

mpls te

Configure the UNI side interface of PTN:

Figure 4.27 : Configure the interfaces to node B

In this scenario the Node B will be connected to the PTN using 4 E1 interfaces and will be working in ATM mode so we have to change the port mode to layer 2

Figure 4.27: Creating the IMA group

We then create the IMA group and then bind the E1 link to it.

Figure 4.28: Enabling the IMA group and setting IMA parameters

After binding the E1 interfaces to the IMA group we set the IMA group parameters to match the Node B then enable the IMA group.

Figure 4.29 : Verify IMA group operation status

If the setting s of the IMA group are correct and at least one E1 link is up the IMA group should be operational.

Configure static routes to reach other equipment in the network:

Figure 4.30 : Configuring static routes on PTN

Since PTN is not running any dynamic IGP protocol with the Node equipment I need to configure static routes to have reachability for loopback IP and interface IP between PTN and all equipment.

Do not forget to configure the static routes on the Node connected to the PTN and import the routes into IGP. Finally do ping tests to verify reachability.

Node connected to PTN configuration:

#

ip route-static 150.1.4.4 255.255.255.255 155.1.24.4

#

import-route static

#

Since the service is enabled on the interface connected to PTN no need to import direct just static.

Remove DCN function from UNI side interfaces:

Figure 4.31 : Disabling DCN from UNI interface

It is very important to disable DCN from the UNI side interfaces or else service configuration will fail.

Creation of MPLS TE tunnels between PTN and NODE:

In this step the following tasks will be accomplished:

1-Creation of main tunnels

2-Creation of protection tunnels

3-Enabling of MPLS OAM function

4-Configuring APS protection group for high availability

Tunnel creation:

Figure 4.32: Configuring the LSR ID of PTN

Make sure LSR ID is configured on the PTN and all NODE equipment and that MPLS and MPLS TE is enabled on all equipment.

Figure 4.33 : Tunnel creation details

When creating the tunnels please follow the below guidelines:

1-The LSP name must be unique on all equipment

2-For PTN I can set Out Interface and In Interface

3-For NODE I should only specify Next hop

4-Allow the U2000 to automatically assign label values this decreases the risk of duplicate label assignment in the whole network

5-For LSP name always use the format Tunnelx/y/z

6-If between the Ingress Node (PTN) and the Egress node (NODE) there are multiple routers make sure to add them and specify their role as transit in the order are in the physical topology.

8

Figure 4.34 : Tunnel creation and transit routers

Figure 4.35 : Verifying tunnel status

The tunnel status ca be seen by checking the following: Service > Tunnel>Manage Tunnels

Configuring MPLS OAM:

Figure 4.36: Configuring MPLS OAM

MPLS OAM is needed for sub second detection of faults in the TE tunnels this can be configured by right clicking on the tunnel in the manage tunnels view and selecting Configure MPLS OAM

For MPLS OAM we have two types of detection:

1-CV which provides detection period of 1000ms (1 second) and cannot be changed

2-FFD which provides sub second detection period which can be configured (recommended)

For MPLS OAM configuration, make sure the OAM packets will go in a circular path using the main tunnels and the protection tunnels.

Example: For traffic going from PTN to NODE: OAM for the forward tunnel will go through the forward tunnel and return on the reverse protection tunnel and vice versa. This is shown in the diagram below:

Figure 4.37: OAM packet path

4.8.2 Configuring APS protection:

There are two main tunnels between the PTN and PE-AGG the forward tunnel and the backward tunnel. Therefore we require two tunnels for protection forward protection tunnel and backward protection tunnel.

It is also important that the protection tunnel path should be different from the main tunnel path but all should begin from the PTN and end in the PE-AGG NODE.

Figure 4.38: Creating protection group

For creating a protection group go to Service > Tunnel > Create Protection group

Make sure I correctly define which tunnel is working and which tunnel is protection.

Configuring ATM service:

I must configure ATM QoS policy to be able to deploy service

To correctly configure the ATM QoS policy it’s important to know what kind of services will be carried and which PVC will carry which service. This information could be obtained from documentation or from the team responsible for deployment of wireless equipment.

An example of the required information:

To create a global ATM service profile go to the Configuration menu and select PTN QoS Profile and then ATM Profile. Make sure I accurately set the parameters for the ATM QoS profile such as SCR and PCR to not exceed the true bandwidth or else errors will happen in configuration.

Figure 4.39 : Adding a new profile

Figure 4.40: Setting profile parameters

Figure 4.41: Finished profile

ATM service creation:

From the U2000 service menu select PWE3 service and select create service:

Figure 4.42 : Service creation

After this I need to configure the source and sink nodes of the ATM PWE3 service as shown below:

Figure 4.43 : Configuration of the Source and Sink Nodes

For example if I set the ID to 1 the U2000 will create ima interface ima1/0/1.1 if I set the ID to 2 the U2000 will create interface ima1/0/1.2 and so on as shown in the figure below:

Figure 4.44: Configuration of the NODE side service

After configuring the source and sink nodes the next step is to configure which PTN will be carried on the new psudowire this can be done by clicking on the ATM Link button as shown in the below figure:

Figure 4.45: Configuring which PVC are carried on the psudowire

Unless I set and ATM QoS policy I will not be able to finish the configuration of the ATM Link.

Finally I will need to bind the new service to an existing MPLS TE tunnel as shown in the figures below:

Figure 4.46 : Selecting the Tunnel that will carry the PWE3 service

Figure 4.47: Final Configuration

Once I have done all configurations and created the service I will need to telnet to the NODE equipment and modify the configuration for the PWE3 to work. The configurations that will be created on the NODE are as below:

interface Ima-group1/0/1.1

pvc 5/41

pvc 5/42

pvc 5/43

mpls static-l2vc destination 150.1.4.4 7 transmit-vpn-label 18 receive-vpn-label 20 tunnel-policy TE

I need to remove the service and create it again however this time I will add the control word function as shown in the below configuration:

interface Ima-group1/0/1.1

pvc 5/41

pvc 5/42

pvc 5/43

mpls static-l2vc destination 150.1.4.4 7 transmit-vpn-label 18 receive-vpn-label 20 tunnel-policy TE control-word

Verification of Service:

Figure 4.48 : Check the status of the service

Figure 4.49: Checking which PVCs are carried by the service

Figure 4.50: Checking service QoS policy and running status

As we can see in the figure above the services are tasted and they are enable and active.

The purpose of this thesis was to present and to configure a network based on MPLS-TP.

After the network was made, the services configured, the tunnels were populated with ATM services. In the picture below we can see that the purpose has been reached, the tunnels are populated successfully with ATM services which are up and enable.

Conclusions

This thesis presents the way to create a packet transport network using MPLS-TP. It is presented that the MPLS-TP protocol is applicable and follows all the requirements to be compatible with MPLS network. Different network scenarios, combining packet and circuit switching properties with MPLS-TP labels, are presented. At the beginning of this thesis, are provided the characteristics and requirements of MPLS-TP protocol which the standardization of this is on going. Furthermore, it is explained how the MPLS-TP management and the forwarding plane work. Some references are also given not only to OAM mechanisms, but also to control plane that the MPLS-TP uses. We use both, global and local significance MPLS-TP labels for configuring the network.

This thesis helped me understand how to use the MPLS-TP protocol that I have installed, configured, and customized for the proposed topology.

In my present work I made my contribution by:

Physical Network Creation in Huawei Lab: Network Design;

Installing systems: setting up boards, creating connections between the equipment;

Creating MPLS-TP tunnels between PTNs

Configuration of packet transmission rules between the equipment: for the protection system (APS)

Customize the system by modifying the dashboards menu, creating specific reports, creating alerts to detect abnormal activity connections between equipment in line with network needs.

Future research directions

Expansion of the number of integrated equipments in the network in order to determine and follow the behavior of the protocol in the extended network

Detect numerous alarms and create troubleshooting procedures

Bibliographu

Cisco Sustems, Understanding MPLS-TP and Its Benefits, White paper,2009. Uuless D. Black, MPLS and Label Switching Networks, pages 5-9, Januaru 2001, Prentice hall PTR, Upper Saddle River, New Jerseu.

Dieter Beller, Rolf Sperber, MPLS-TP – The New Technologu for Packet Transport Networks, Alcatel-Lucent Deutschland AG.

James E. Goldman & Phillip T. Rawles, Applied Data Communications, 2004 (ISBN 0-471-34640-3)

RFC 3031, Multiprotocol Label Switching Architecture, Januaru 2001

harru G. Perros, Connection-oriented Networks SONET/SDh, ATM, MPLS and Optical Networks, John Wileu & Sons Ltd, 2005 (ISBN 0-470-02163-2)

T-PACK, T-MPLS, a new route to carrier Ethernet, June 2007

MPLS Transport Profile -7750 SR Advanced Configuration Guide

ITU-T Recommendation G.812: “Timing requirements of slave clocks suitable for use as node clocks in synchronization networks”, June 2004

ITU-T Recommendation G.8262: “Timing characteristics of synchronous ethernet equipment slave clock (EEC)”, August 2007

RFC 3916, Requirements for Pseudo-Wire Emulation Edge-to-Edge (PWE3),

M. Bocci, Ed. Alcatel-Lucent S. Bruant, Ed. D. Frost, Ed. Cisco Sustems L. Levrau, Alcatel-Lucent L. Berger, LabN (IETF MPLS Working Group), A Framework for MPLS in Transport Networks, Februaru

Xiao, X., McPherson, D., and Pate, P.: “Requirements for Pseudo-Wire Emulation Edge-to-Edge (PWE3)”, IETF RFC3916, September 2004

B. Niven-Jenkins, Ed. BT D. Brungard, Ed. AT&T M. Betts, Ed. huawei Technologies N. Sprecher Nokia Siemens Networks S. Ueno NTT Communications, (RFC 5654) Requirements of an MPLS Transport Profile, September 2009

S. Mansfield, Ed. E. Grau, Ed. Ericsson h. Lam, Ed. Alcatel-Lucent, MPLSTP Network Management Framework, Januaru 2010

hing-Kam Lam, Alcatel-Lucent Scott Mansfield, Eric Grau Ericsson, MPLSTP Network Management Requirements, October 2009

Cisco Sustems, Cisco MPLS Controller Software Configuration Guide, April 2000

ITU Recommendation G.709/U.1331, Interfaces for the Optical Transport Network (OTN), March 2003

MPLS Transport Profile(MPLST-TP) – A set of enhancements of Rich MPLS Too, Juniper Networks 2011

MPLS Transport Profile- Test Methodologies-IXIA November 2010

Packet Transport Networks: Overview and Future Direction 26 August 2014, Bangkok, Thailand

ITU-T Recommendation G.783: “Characteristics of synchronous digital hierarchy (SDH) equipment functional blocks”, March 2003

ITU-T Recommendation G.811: “Timing Characteristics of Primary Reference Clocks”, September 1997

ITU-T Recommendation G.813: “Timing characteristics of SDH equipment slave clocks (SEC)”, March 2003

http://support.huawei.com/enterprise/en/network-management/imanager-u2000-pid-15315

http://opti500.cian-erc.org/opti500/pdf/sm/mpls-tp%20cisco.pdf

ITU-T Recommendation M.3010: “Principles for a telecommunications management network”, February 2000 [M.3400] ITU-T Recommendation M.3400: “TMN management functions”, February 2000