Đề tài QoS over MPLS for Hutech network

Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 1 - Table of Contents Architecture of Subject.......................................................................................4 Relation Words ...................................................................................................5 Multiprotocol label switching ............................................................................6 Actuality of MPLS at VietNam..........................................................................6 Advantage of MPLS...........................................................................................8 Disadvantage of MPLS.......................................................................................8 Icon use in subject ..............................................................................................9 CHAPTER 1: INTRODUCTION ABOUT NETWORK SYSTEM OF HUTECH UNIVERSITY .................................... 10 1.1. Description about network system of Hutech University. ..........................11 1.2. Important problem and solution. .................................................................11 1.3. Task of subject.............................................................................................12 CHAPTER 2: QOS OVER MPLS NETWORK.........................13 Part 1: Overview about MPLS .................................................... 14 2.1. Architecture of MPLS packet......................................................................15 2.1.1. Label. ..................................................................................................15 2.1.2. Experimental. ......................................................................................15 2.1.3. Bottom of Stack. .................................................................................15 2.1.4. Time to Live........................................................................................16 2.2. Operating of MPLS network. ......................................................................16 2.2.1. MPLS domain. ....................................................................................16 2.2.2. Ingress and egress node. .....................................................................17 2.2.3. Label Switch Router. ..........................................................................17 2.2.4. Label Switch Path. ..............................................................................18 2.2.5. Forwarding Equivalent Class..............................................................18 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 2 - 2.2.6. Label Distribution Protocol. ...............................................................19 2.3. Command for configure MPLS...................................................................20 Part 2: Overview about QoS ....................................................... 22 2.4. Architecture of QoS.....................................................................................23 2.4.1. IntServ model......................................................................................23 2.4.2. DiffServ model....................................................................................25 2.4.3. Different between IntServ model and DiffServ model.......................27 2.5. Classification. ..............................................................................................27 2.6. Marking. ......................................................................................................27 2.7. Queuing tools...............................................................................................28 2.7.1. First In-First Out Queuing. .................................................................29 2.7.2. Priority Queuing..................................................................................31 2.7.3. Custom Queuing. ................................................................................32 2.7.4. Weighted Fair Queuing.......................................................................33 2.7.5. Class-Based Weighted Fair Queuing. .................................................38 2.7.6. Low-latency Queuing. ........................................................................41 Part 3: QoS over MPLS...............................................................44 2.8. Relation about IPP, DSCP and MPLS EXP. ...............................................45 2.8.1. IPP.......................................................................................................45 2.8.2. DSCP...................................................................................................46 2.8.3. MPLS EXP..........................................................................................47 2.9. DiffServ with IP packets. ............................................................................48 2.10. DiffServ with MPLS packets. ...................................................................50 2.11. DiffServ Tunneling Modes for MPLS networks.......................................52 2.11.1. Pipe Model. .......................................................................................52 2.11.2. Short-Pipe Model. .............................................................................54 2.11.3. Uniform Model. ................................................................................55 2.12. Steps implement QoS over MPLS.............................................................57 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 3 - CHAPTER 3: NETWORK DESIGN AND IMPLEMENT ....... 58 3.1. Building solution for Hutech network.........................................................60 3.1.1. Real model of Hutech network. ..........................................................60 3.1.2. Solution model for Hutech network....................................................61 3.2. Building simulation model to resolve for Hutech network. ........................62 3.2.1. Simulation model. ...............................................................................62 3.2.2. Implement QoS over MPLS in simulation model. .............................64 3.3. Get Result. ...................................................................................................72 Get Result and Define of develop in Subject..............................74 References .........................................................................................................75 Index ...................................................................................................................76 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 4 - Architecture of subject Subject includes three chapters: Chapter 1: Introduction about network system of Hutech University, problem of Hutech network system, solution to resolve. Chapter 2: Chapter 2 includes three parts. Part 1: Overview about MPLS, architecture MPLS packet, operation of MPLS network, command line for configure MPLS operation. Part 2: Overview about QoS, architecture of QoS, classification, marking and queuing tool. Part 3: QoS over MPLS, relative about IPP, DSCP and MPLS EXP, DiffServ with IP packet and MPLS packet, DiffServ tunneling mode, steps implement QoS over MPLS network. Chapter 3: Network design and implement. Deploy QoS over MPLS, get result and define of develop in subject. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 5 - Relation Words IPP : IP Precedence (value support implement QoS) DSCP : Differentiated Services Code Point MPLS : Multiprotocol Label Switching EXP : Experimental QoS : Quality of Service LSP : Label Switched Path LSR : Label Switched Router IntServ :Iintegrated services DiffServ : Differentiated Services LLQ : Low-latency Queuing FIFO : First In – First Out CQ : Custom Queuing WFQ : Weighted Fair Queuing CBWFQ : Class-Based Weighted Fair Queuing LDP : Label Distribution Protocol Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 6 - Multiprotocol Label Switching Multiprotocol Label Switching (MPLS) is a standards-approved technology for speeding up network traffic flow and making it easier to manage. MPLS involves setting up a specific path for a given sequence of packets, identified by a label put in each packet, thus saving the time needed for a router to look up the address to the next node to forward the packet to. MPLS is called multiprotocol because it works with the Internet Protocol (IP), Asynchronous Transport Mode (ATM), and frame relay network protocols. With reference to the standard model for a network (the Open Systems Interconnection, or OSI model), MPLS allows most packets to be forwarded at the layer 2 (switching) level rather than at the layer 3 (routing) level. In addition to moving traffic faster overall, MPLS makes it easy to manage a network for quality of service (QoS). For these reasons, the technique is expected to be readily adopted as networks begin to carry more and different mixtures of traffic. Actuality of MPLS at VietNam -With VietNam, MPLS deployment are building in communication network at VNPT head of company VietNam. With VoIP project is deploying, VNPT established an axle MPLS network with three LSR core. LSRs edge will be invested and extended at places have large demand as Hai Phong, Quang Ninh at north, Da Nang, Khanh Hoa…at medium, Binh Duong, Dong Nai, Ba Ria – Vung Tau…at south. Next, FPT Telecom, Viettel, electricity are taked part and create competition environment with high QoS and cheap. - Present, not including companies and foreign office representative, there are a lot of home companies in finance field, insurance, bank use this service (Bao Viet insurance company, Dong A bank…). Beside that, arrange state as Ministry of Finance, customs, treasury, tax associated together by VPN/MPLS. -VPN/MPLS technology officially deployed, applied and test successfully and inserted to exploited from 2003 by VDC. 2004, VPN MPLS solution of VDC Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 7 - saved up technology information IT Week 14 gold cup and extended to exploited on all 64 provinces of all the country with trade name VPN/VNN. - VPN/VNN MPLS solution of VDC apply and deploy to rely on pass and device technology of Cisco, with target create a network solution safe, security, slow late and intergrate with each apply as Data, Voice, Video… Price for first setup: include price for setup and link to MPLS/VNN Order Speed Price for channel TDNH (VND/channel/time) Price for setup, link to network VPN/VNN(VND/ch annel/time) 1 64 Kbps 1,500,000 2,000,000 2 128Kbps<speed <=896kbps 5,000,000 2,000,000 3 1Mbps<speed<= 2Mbps 5,000,000 3,000,000 4 2Mbps<speed<= 10Mbps 20,000,000 5,000,000 5 10Mbps<speed<=155Mbps 20,000,000 10,000,000 Price for month: (Upcountry price + Service price MPLS/VNN) Order Speed Price for channel TDNH (VND/port /month) Price for all (VND/port /month) 1 64Kbps 609,000 958,000 2 128Kbps 875,000 1,447,000 3 192Kbps 1,104,000 1,694,000 4 156Kbps 1,370,000 1,941,000 5 384Kbps 1,705,000 2,415,000 6 512Kbps 2,114,000 2,994,000 7 768Kbps 2,600,000 3,558,000 8 896Kbps 2,886,000 3,682,000 9 1024Kbps 3,171,000 3,928,000 10 1536Kbps 4,394,000 5,442,000 11 2048Kbps 5,112,000 5,978,000 12 4Mbps 10,224,000 11,561,000 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 8 - 13 6Mbps 15,758,000 14 8Mbps 18,914,400 20,529,000 15 10Mbps 25,661,000 16 34Mbps 24,536,000 33,034,000 17 45Mbps 42,173,000 63,964,000 Advantage of MPLS: -Security (absolute security in core MPLS network and local loop network). -Flexible (Easy for wire-open). -Easy for administrator control. Disadvantage of MPLS: -VietNam has limit human resource for control MPLS network system. -Device support MPLS network is limit. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 9 - Icon use in subject Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 10 - CHAPTER 1: INTRODUCTION ABOUT NETWORK SYSTEM OF HUTECH UNIVERSITY Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 11 - 1.1. Description about network system of Hutech University. Hutech University include three branches, branch 1 at Binh Thanh distinct, branch 2 at Phu Nhuan distinct, and the last branch at Thu Duc distinct. Three branches join together by Frame-Relay technology. In every branch, include type of faculty: -Faculty of information technology. -Faculty of economy. -Faculty of build. -Faculty of foreign language. -Faculty of electron. In addition, every branch has SQL server, Web Server, FTP Server, Mail Server. Network diagram Figure 1.1- Network diagram of Hutech University. 1.2. Important problem and solution. Hutech University is carrying to enlarge infrastructure. Up to this time, the number of branch and department are growing a lot. For this reason, network system of Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 12 - Hutech University is old and stunted. This network system can’t satisfy need to communication information between branches. Network system has limit bandwidth and old technology, so obstruct at any time in network system. To make good that problem, solution for Hutech University network system must a new technology, high effect, low cost. We have too much solutions, and the best solution is QoS over MPLS. 1.3. Task of Subject. With Hutech’s problem, we will build technology QoS over MPLS for Hutech network. We design and implement to preference for important traffic, example video traffic, voice traffic… To limit obstruct. Model solution for Hutech network: Figure 1.2 – Solution model for Hutech network Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 13 - CHAPTER 2: QOS OVER MPLS NETWORK Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 14 - PART 1: OVERVIEW ABOUT MPLS Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 15 - 2.1. Architecture of MPLS packet. MPLS label is a field with 32 bit with hard architecture. Figure 2.1 - MPLS label architecture. 2.1.1. Label. Label include 20 bit in header MPLS, their value between 0 to 220–1 and it has 1,048,575 labels. However, 16 value labels first aren’t use. In IP network, transport packets must use IP source and IP destination, but with MPLS network, packets are transport by label. Routers use label instead for IP address. 2.1.2. Experimental. From bit 20 to bit 22 is exp field, three bits used for quality of services. Exp bit similar Precedence bit in IP header. In IP network, implement quality of service is use IP Precedence or DSCP, but with MPLS network is use Experimental. 2.1.3. Bottom of Stack. Bit 23 BoS (Bottom of Stack) in MPLS header, if label is bottom of stack, it has value 1, if label isn’t bottom of stack it has value 0. Figure 2.2 – Label of Stack. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 16 - In MPLS network, when transport packets, every router in MPLS network is use label for forward packets to exactly destination, router is use label table. 2.1.4. Time to Live. From bit 24 to bit 31 are used for TTL (Time to Live) field. This field similar TTL field in IP header. If router can’t finds the destination of packet but router is forwarding that packet, this action will loop. So TTL field make avoid look. When packet through every router it’s drop 1. When TTL field has value 0, router will drop it. 2.2. Operating of MPLS network. 2.2.1. MPLS domain. MPLS domain include two parts: -Core network (core). -Edge network (edge). With cord network, core network operate complete in MPLS network, router operate in core network will assign label to packet and forward that packet to next router. With edge network, router in edge network must do two tasks; two tasks are imposition label and disposition label from packet. In case imposition with packet through from IP network to MPLS network, packet is imposition label and that operation is call imposition. With case packet through from MPLS network to IP network, packet is disposition label and operation is call disposition. Figure 2.3 - Imposition and Disposition. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 17 - 2.2.2. Ingress and Egress node. When packet goes from IP network to MPLS network, packet will assign label. And operation is call ingress. When packet goes from MPLS network to IP network, packet will unassigned label. And operation is call egress. Both ingress and egress node are edge router. 2.2.3. Label Switch Router. A label switch router (LSR) is a router that supports MPLS. It is capable of understanding MPLS labels and of receiving and transmitting a labeled packet on a data link. Three kinds of LSRs exist in an MPLS network: Ingress LSRs receive a packet that is not labeled yet, insert a label (stack) in front of the packet, and send it on a data link. Egress LSRs receive labeled packets, remove the label(s), and send them on a data link. Ingress and egress LSRs are edge LSRs. Intermediate LSRs receive an incoming labeled packet, perform an operation on it, switch the packet, and send the packet on the correct data link. An LSR can do the three operations: pop, push, or swap. It must be able to pop one or more labels (remove one or more labels from the top of the label stack) before switching the packet out. An LSR must also be able to push one or more labels onto the received packet. If the received packet is already labeled, the LSR pushes one or more labels onto the label stack and switches out the packet. If the packet is not labeled yet, the LSR creates a label stack and pushes it onto the packet. An LSR must also be able to swap a label. This simply means that when a labeled packet is received, the top label of the label stack is swapped with a new label and the packet is switched on the outgoing data link. An LSR that pushes labels onto a packet that was not labeled yet is called an imposing LSR because it is the first LSR to impose labels onto the packet. One that is doing imposition is an ingress LSR. An LSR that removes all labels from Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 18 - the labeled packet before switching out the packet is a disposing LSR. One that does disposition is an egress LSR. 2.2.4. Label Switch Path. A label switched path (LSP) is a sequence of LSRs that switch a labeled packet through an MPLS network or part of an MPLS network. Basically, the LSP is the path through the MPLS network or a part of it that packets take. The first LSR of an LSP is the ingress LSR for that LSP, whereas the last LSR of the LSP is the egress LSR. All the LSRs in between the ingress and egress LSRs are the intermediate LSRs. Figure 2.4- Label Switch Path 2.2.5. Forwarding Equivalent Class. A Forwarding Equivalence Class (FEC) is a group or flow of packets that are forwarded along the same path and are treated the same with regard to the forwarding treatment. All packets belonging to the same FEC have the same label. However, not all packets that have the same label belong to the same FEC, because their EXP values might differ; the forwarding treatment could be different, and they could belong to a different FEC. The router that decides which packets belong to which FEC is the ingress LSR. This is logical because the ingress LSR classifies and labels the packets. Following are some examples of FECs: -Packets with Layer 3 destination IP addresses matching a certain prefix. -Multicast packets belonging to a certain group Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 19 - -Packets with the same forwarding treatment, based on the precedence or IP DiffServ Code Point (DSCP) field. -Layer 2 frames carried across an MPLS network received on one VC or (sub)interface on the ingress LSR and transmitted on one VC or (sub)interface on the egress LSR. -Packets with Layer 3 destination IP addresses that belong to a set of Border Gateway Protocol (BGP) prefixes, all with the same BGP next hop. This last example of a FEC is a particularly interesting one. All packets on the ingress LSR for which the destination IP address points to a set of BGP routes in the routing table—all with the same BGP next-hop address—belong to one FEC. It means that all packets that enter the MPLS network get a label depending on what the BGP next hop is. 2.2.6. Label Distribution Protocol. To get packets across a label switched path (LSP) through the MPLS network, all LSRs must run a label distribution protocol and exchange label bindings. When all the LSRs have the labels for a particular Forwarding Equivalence Class (FEC), the packets can be forwarded on the LSP by means of label switching the packets at each LSR. The label operation (swap, push, pop) is known to each LSR by looking into the LFIB. The LFIB (which is the table that forwards labeled packets) is fed by the label bindings found in the LIB. The LIB is fed by the label bindings received by LDP, Resource Reservation Protocol (RSVP), MP-BGP, or statically assigned label bindings. Because RSVP distributes the labels only for MPLS traffic engineering and MP-BGP distributes the labels only for BGP routes, you are left with LDP for distributing all the labels for interior routes. Therefore, all directly connected LSRs must establish an LDP peer relationship or LDP session between them. The LDP peers exchange the label mapping messages across this LDP session. A label mapping or binding is a label that is bound to a FEC. The Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 20 - FEC is the set of packets that are mapped to a certain LSP and are forwarded over that LSP through the MPLS network. LDP has four major functions: -The discovery of LSRs that are running LDP -Session establishment and maintenance -Advertising of label mappings -Housekeeping by means of notification When two LSRs are running LDP and they share one or more links between them, they should discover each other by means of Hello messages. The second step is for them to establish a session across a TCP connection. Across this TCP connection, LDP advertises the label mapping messages between the two LDP peers. These label mapping messages are used to advertise, change, or retract label bindings. LDP provides the means to notify the LDP neighbor of some advisory and error messages by sending notification messages. 2.3. Command for configure MPLS. Command used for configure MPLS operate Command used for verify MPLS operate Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 21 - Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 22 - PART 2: OVERVIEW ABOUT QOS. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 23 - 2.4. Architectures of QoS. There are three part for enforce QoS : -QoS in a environment single network (as arrange sequence, make lists sequencing and tools for transmit information on network). -The technique transmit signal for regulate QoS among factors into network. -The policy QoS, administer, and calculate features for control and manage transmit information among nodes into network. 2.4.1. IntServ model. Integrated services (IntServ) defines a different model for QoS than does DiffServ. IntServ defines a signaling process by which an individual flow can request that the network reserve the bandwidth and delay needed for the flow. The original work grew out of the experiences of the IETF in multicasting the audio and video for IETF meetings in the early to mid-1990s. Figure 2.5- IntServ model. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 24 - IntServ admission control decides when a reservation request should be rejected. If all requests were accepted, eventually too much traffic would perhaps be introduced into the network, and none of the flows would get the requested service. IntServ uses Resource Reservation Protocol for signaling to reserve the bandwidth. With a full IntServ implementation (more on that later), the originator of the flow (Hannah) begins signaling. At each router along the route, the router asks itself, “Can I support this request?” If the answer is yes, it forwards the request to the next router. Each router holds the bandwidth temporarily, waiting on the confirmation to flow back to the originator (Hannah). When each router sees the reserve RSVP command flow back to the originator, each router completes the reservation. What does it mean for the router to “reserve” something? In effect, the router reserves the correct queuing preferences for the flow, such that the appropriate amount of bandwidth is allocated to the flow by the queuing tool. RSVP can also request a certain (low) amount of delay, but implementing a guarantee for delay is a little more difficult; IOS, for instance, just reserves the queuing preference. In fact, IntServ RFCs actually define the term “guarantee” as a relatively loose goal, and it is up to the actual implementation to decide how rigorous or general to make the guarantees. RSVP continues signaling for the entire duration of the flow. If the network changes, or links fail and routing convergence occurs, the network may no longer be able to support the reservation. Therefore, RSVP reserves the bandwidth when the flow initializes and continues to ensure that the flow can receive the necessary amount of bandwidth. IntServ has some obvious disadvantages, and it has several advantages. IntServ actually predates DiffServ; DiffServ, to some degree, was developed to provide an Internet-scale QoS model, because IntServ scales poorly. IntServ expects the hosts to signal for service guarantees, which brings up two issues—whether the hosts can be trusted by the network and whether the hosts actually support RSVP. Alternatively, routers can be configured to reserve bandwidth on behalf of hosts, but the configuration can quickly become an administrative problem because Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 25 - additional configuration would need to be added for each reserved flow. Also IntServ works best when all intermediate networks support IntServ. 2.4.2. DiffServ model. The DiffServ model designed to repair limits of IntServ model. The DiffServ model can flexible high and extend large. Instead of perform through QoS and unity on all line as IntServ model, the Diffserv model perform QoS individually on each router, so DiffServ unnecessary signal to follow each flow therefore economize bandwidth and can extend, approprivate with large network model. Salient features in manage resources of DiffServ model implemented at: -The DiffServ model don’t implement to signal, shake hand when establish flow therefore it is losed bandwidth for signal. -The DiffServ model manage resource effectly because it don’t reserve resources for any of a services. Services devided follow sequence priority, which service has priority higher will provided resource at regime better, when haven’t flow, the resource will be returned for system and used by other services. Activity of DiffServ Activity of DiffServ can describle as follows: First, information packages classified become a lot of priority group from low to high according to feature of each service, device will provide authority used resource more priority, resource will used by lower group if higher group don’t use. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 26 - Video packet Voice packet Data packet Figure 2.6 - DiffServ Model Solution QoS follow DiffServ performed follow steps: Mark and classify package. First, packages will marked for differentiate, after that arranged in group conformably. Mark and arrange will helf for perform QoS at after steps: -Manage obstructed: Structure manage obstruct to performed on interfaces of network device. When package come to these interfaces, package will classified follow each queue rely on priority. -Avoid obstructed: Structure reject package before obstructe. -Put threshold: Structure put upper threshold, under threshold for bandwidth, specific is bandwidth will ensured a under threshold minimum and when larger than upper threshold package can be rejected or move to queue. -Press header: Header hold large part in a package but don’t have real information, structure press header will economize bandwidth. - Fragmentate: data packages often have large length, This event will cause late and Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 27 - obstructed. Structure fragmentate will mince these packages become smaller packages for avoid obstructed. 2.4.3.Difference between InServ model and DiffServ model. IntServ DiffServ Use signal protocol RSVP for fight resource Don’t use protocol Use for small network and little flow network Use for large network and small network, have a lot of flow network Resource wasted high Resource wasted little Don’t manage resource because marked before that Manage resource to rely on priority of each flow Tabel 2.1- Compare IntServ model and DiffServ model 2.5. Classification. Almost every QoS tool uses classification to some degree. To put one packet into a different queue than another packet, the IOS must somehow differentiate between the two packets. To perform header compression on Real Time Protocol (RTP) packets, but not on other packets, the IOS must determine which packets have RTP headers. To shape data traffic going into a Frame Relay network, so that the voice traffic gets enough bandwidth, the IOS must differentiate between Voice over IP (VoIP) and data packets. If an IOS QoS feature needs to treat two packets differently, you must use classification. Because most QoS tools need to differentiate between packets, most QoS tools have classification features. In fact, many of you will already know something about several of the QoS tools Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 28 - described in this book, and you will realize that you already know how to perform classification using some of those tools. For instance, many QoS tools enable you to classify using access control lists (ACLs). If ACL 101 permits a packet, a queuing tool might put the packet into one queue; if ACL 102 permits a packet, it is placed in a second queue; and so on. In one way of thinking, queuing could instead be called classification and queuing, because the queuing feature must somehow decide which packets end up in each queue. Similarly, traffic shaping could be called classification and traffic shaping, policing could be called classification and policing, and so on. Because most QoS tools classify traffic, however, the names of most QoS tools never evolved to mention the classification function of the tool. Most classification and marking tools, like the other types of QoS tools, generally operate on packets that are entering or exiting an interface. The logic works something like an ACL, but the action is marking, as opposed to allowing or denying (dropping) a packet. More generally, classification and marking logic for ingress packets can be described as follows: -For packets entering an interface, if they match criteria 1, mark a field with a value. -If the packet was not matched, compare it to criteria 2, and then mark a potentially different field with a potentially different value. -Keep looking for a match of the packet, until it is matched, or until the classification logic is complete. 2.6. Marking. Marking accept network devices classify package or frame rely on gait specific description flow. Some gait description flow are used for mark as: class of service (CoS), DSCP, IP priority, MPLS EXP bit, group QoS. Marking is used to establish information in heading package class two or class three. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 29 - Mark package or frame and classify accept network device discriminate easily packages or frame marked. Marking is element useful because it accept network device recognize easily packages or frames follow specific classes. Then QoS technical can apply compatible for right ensure with manage QoS policies. Marking include organize some bits inner a data-link class or network header with purpose helpful for QoS tools of other device can classify rely on some value marked. We can mark a lot of field correlative for each specific request. Some field are used a lot of, other field are not. Some choose inner mark to grill with all device inner local network while other ones only use on base hardware default. And making on WAN same. 2.7. Queuing tools. We have one way for control information overflow, the way is use algorithm queue for arrange traffic and determine some methods for decentralization priority of traffic. IOS of Cisco support some tool following: - First-in, first-out (FIFO) - Priority queuing (PQ) - Custom queuing (CQ) - Weighted fair queuing (WFQ) - Low Latency Queuing (LLQ) Every algorithm was design for solve problem when transmit messages in network, and it is effect for network. 2.7.1. First In – First Out Queuing. The first reason that a router or switch needs output queues is to hold a packet while waiting for the interface to become available for sending the packet. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 30 - Whereas the other queuing tools in this chapter also perform other functions, like reordering packets, FIFO Queuing just provides a means to hold packets while they are waiting to exit an interface. FIFO Queuing does not need the two most interesting features of the other queuing tools, namely classification and scheduling. FIFO Queuing uses a single queue for the interface. Because there is only one queue, there is no need for classification to decide the queue into which the packet should be placed. Also there is no need for scheduling logic to pick which queue from which to take the next packet. The only really interesting part of FIFO Queuing is the queue length, which is configurable, and how the queue length affects delay and loss. FIFO Queuing uses tail drop to decide when to drop or enqueue packets. If you configure a longer FIFO queue, more packets can be in the queue, which means that the queue will be less likely to fill. If the queue is less likely to fill, fewer packets will be dropped. However, with a longer queue, packets may experience more delay and jitter. With a shorter queue, less delay occurs, but the single FIFO queue fills more quickly, which in turn causes more tail drops of new packets. These facts are true for any queuing method, including FIFO. Figure 2.7 – FIFO Queue. Queue has three packets 4, 3, 2, 1, if follow queue packet 1 can pass first and next are three packets 2, 3, 4. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 31 - 2.7.2. Priority Queuing. Priority Queuing’s most distinctive feature is its scheduler. PQ schedules traffic such that the higher-priority queues always get serviced, with the side effect of starving the lower-priority queues. With a maximum of four queues, called High, Medium, Normal, and Low, the complete logic of the scheduler can be easily represented. The PQ scheduler has some obvious benefits and drawbacks. Packets in the High queue can claim 100 percent of the link bandwidth, with minimal delay, and minimal jitter. The lower queues suffer, however. In fact, when congested, packets in the lower queues take significantly longer to be serviced than under lighter loads. In fact, when the link is congested, user applications may stop working if their packets are placed into lower-priority queues. Most of the rest of the details about PQ can be easily understood. PQ classifies packets based on the content of the packet headers. It uses a maximum of four queues, as mentioned earlier. The only drop policy is tail drop—in other words, after classifying the packet, if the appropriate queue is full, the packet is dropped. The length of each queue, which of course affects packet loss and delay, can be changed—in fact, PQ can set the queue length to a value of zero, which means the queue length is infinite. Figure 2.8 – Priority Queue. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 32 - PQ works great for QoS policies that need to treat one type of traffic with the absolute best service possible. However, PQ’s service for the lower queues degrades quickly, making PQ impractical for most applications today. For instance, even running one FTP connection, one web browser, one NetMeeting call, and two VoIP calls when creating the output for this section of the book, the TCP connections for the FTP and HTTP traffic frequently timed out. 2.7.3. Custom Queuing. Custom Queuing (CQ) followed PQ. CQ addresses the biggest drawback of PQ by providing a queuing tool that does service all queues, even during times of congestion. It has 16 queues available, implying 16 classification categories, which is plenty for most applications. The negative part of CQ, as compared to PQ, is that CQ’s scheduler does not have an option to always service one queue first (like PQ’s High queue) so CQ does not provide great service for delay- and jitter-sensitive traffic. CQ was design for some applications or associates, and CQ can share information of network with different applications by small traffic and delay time can agree. In environment like that, bandwidth must be balance with every application and user. If we use particularity of CQ algorithm support by Cisco for make sure about bandwidth, where in the network has obstructed network, or make sure for transmit information will be ok with bandwidth we issue and establish. Requests of guest will be arrange by set up some tools, size of queue with every class of packet and process of packet are use round-robin algorithm. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 33 - Figure 2.9 – Classification and move packet into SNA queue. In figure 3.7.3, packing of SNA system need ensure some small amount of service. We can provide a haft of bandwidth for transmit data in SNA system, and we issue remain bandwidth for another protocol, example IP or IPX. Algorithm will arrange messages into 1 of 17 queues (queue 0 store message of system of system, example test keepalive always send update connection….). Router will do that, arrange information into queue form queue 1 to queue 16, and router use round- robin algorithm, arrange every byte. That function is make sure not application can operate can use resource highest than system issue. Similar PQ, CQ is configuring static and don’t automatic update if network has change. 2.7.4. Weighted Fair Queuing. Weighted Fair Queuing differs from PQ and CQ in several significant ways. The first and most obvious difference is that WFQ does not allow classification options to be configured! WFQ classifies packets based on flows. A flow consists of all packets that have the same source and destination IP address, and the same source and destination port numbers. So, no explicit matching is configured. The other large difference between WFQ versus PQ and CQ is the scheduler, which simply Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 34 - favors low-volume, higher-precedence flows over large-volume, lower-precedence flows. Also because WFQ is flow based, and each flow uses a different queue. Flow-Based WFQ, or just WFQ, classifies traffic into flows. Flows are identified by at least five items in an IP packet: -IP source. -IP destination. -Transport protocol. -TCP source port or UDP source port. -TCP destination port or UDP destination port. -Value Precedence of IP packet. Because WFQ classification packet to rely on row of different traffic and then it move that row into different queue, router has total different queues. These queues more than every different queue tools. WFQ use algorithm different with algorithm every queue tools, that different is control more traffic. However, WFQ can describe like this: -Every rows have the same priority of packet will be have the same bandwidth, and it don’t care rows have how many byte in every traffic row. -If row has different precedence, if the row has precedence highest with has bandwidth high. -Finally, WFQ will priority for rows have traffic small and priority high. Example, if WFQ is controlling 10 queues with different IP Precedence in port has 128 kbps; every traffic row can have 12.8 kbps, so delay time will be big. And goal second of WFQ is provide enough bandwidth for traffic rows have high precedence. For do that, rows issue number IP Precedence + 1. Example, traffic has IP Precedence has value 7, and this traffic has bandwidth high bandwidth of IP Precedence has value 0. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 35 - Figure 2.10 – Model of operation WFQ Time process WFQ To get purpose issue bandwidth, WFQ use regulator time so simple. Regulator time use packet has one after another index low sequence number, and it call SN, when it transmit packet in the next hardware queue. Mechanism WFQ issues every packet with one after another index SN when packet goes into WFQ queue. Process issue one after another SN is a part important in mechanism regulator time of WFQ. Regulator time WFQ calculates one after another numbers SN by parameters of flow traffic, include length and priority of packet. Syntax to calculate one after another number SN of packet in flow traffic like this: SN= SN number before +( weight*length of new packet). Weight like this: Weight=32384 / (IP_Precedence+1) Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 36 - Syntax reference to length of new packet, weight of flow traffic and value of SN before. By consider length of packet, and calculator one after another SN number may be has a high SN number for packets, and packets have big size and one after another SN number low more than packets have small size. Include one after another SN number of packet before it moves in queue, syntax will calculator and result is a biggest SN number than packets in queue has biggest packet. It will issue (IPP+1), packets have high priority, and it will have low SN. In figure 2.11, describe how two packets issue two one after another. Calculator one after another number so easy. However, the first packet in a flow traffic is don’t has one after number SN of the first packet if it was use that syntax. Syntax describe one after another SN of the end packet will move into hardware queue, and it use one after number SN follow a new next flow. Figure 2.11- Describe calculator SN (Sequence Number) After one after number SN was issue, next work chooses which the packet will be removed in regulator time device. It will take packet has low SN in queue. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 37 - Policy to reject packet of WFQ, number and length of queue. In operator of router, although traffic match in queue, but if traffic through router still crush, router must reject packet to avoid obstructed. WFQ use second process and call modified tail drop for choose what packet will reject. -First, WFQ will consider the best limit of all packets in queue, and limit call is hold-queue limit. If packet goes to queue and hold-queue is limit, packet will be drop. That decide don’t belong to one queue, it belong to queue system of WFQ. Different way, hold-queue limit is a local number, and calculates by total WFQ queue. -Second, WFQ consider length of queue and packets are move in queue. Before packets move into queue, congestive discard threshold will test with the true length of queue. If the length of queue longer than CDT, packet will be drop, but new packet will not drop. Packet with one after another number in queues of WFQ will be drop. Figure 2.12 – Describe process of WFQ The hold-queue size limits the total number of packets in all of the flow or conversation queues. However, CDT limits the number of packets in each individual queue. If CDT packets are already in the queue into which a packet Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 38 - should be placed, WFQ considers discarding the new packet. Normally, the new packet is discarded. If a packet with a larger SN has already been enqueued in a different queue, however, WFQ instead discards the packet with the larger SN! It’s like going to Disneyland, getting in line, and then being told that a bunch of VIPs showed up, so you cannot ride the ride, and you will have to come back later. (Hopefully Disney would not take you out of the line and send you to the bit bucket, though!) In short, WFQ can discard a packet in another flow when the queue for a different flow has exceeded CDT but still has lower sequence numbers. You can configure the CDT to a value between 1 and 4096, inclusive. Finally, WFQ can be configured for a maximum of 4096 queues, but interestingly, the actual value can only be a power of 2 between 16 and 4096, inclusive. The IOS restricts the values because WFQ performs a hash algorithm to classify traffic, and the hash algorithm only works when the number of queues is one of these valid values. 2.7.5. Class-Based Weighted Fair Queuing. Like the other queuing tools with WFQ in the name, CBWFQ uses features that are similar to some other queuing tools, and completely different from others. CBWFQ is like CQ, in that it can be used to reserve minimum bandwidth for each queue, but it differs from CQ in that you can configure the actual percentage of traffic, rather than a byte count. CBWFQ is like WFQ in that CBWFQ can actually use WFQ inside one particular queue, but it differs from WFQ in that it does not keep up with flows for all the traffic. Many people find it difficult to keep the details memorized. To help overcome confusion, the features of CBWFQ are covered in the next several pages. At the end of this section, some summary tables list the key features and compare CBWFQ to some of the other queuing tools. CBWFQ supports 64 queues, with a maximum and default queue length of 64. All 64 queues can be configured, but one class queue, called class-default, is automatically configured. If the explicitly configured classification does not match Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 39 - a packet, IOS places the packet into the class-default class. You are allowed to change the configuration details regarding this default class, but this one class always exists. CBWFQ provides a great advantage by allowing WFQ to be used in the class- default queue. You may recall that WFQ is actually a very good default choice for queuing, because it treats low-volume flows well, and many low-volume flows are also interactive flows. WFQ also treats packets with high precedence well. So, with CBWFQ, for the traffic you know about, you classify it, and reserve the right amount of bandwidth for the class. For the traffic you cannot characterize, you let it default into the class-default queue, where you can dynamically apply some fairness to the default traffic by using WFQ. The capability to reserve bandwidth for some packets, and fairly assign the rest of the bandwidth with WFQ, makes CBWFQ a very powerful queuing tool. Cisco creates CBWFQ and LLQ, CBWFQ uses features that are similar to some other queuing tools, and completely different from others. CBWFQ is like CQ, in that it can be used to reserve minimum bandwidth for each queue, but it differs from CQ in that you can configure the actual percentage of traffic, rather than a byte count. CBWFQ is like WFQ in that CBWFQ can actually use WFQ inside one particular queue, but it differs from WFQ in that it does not keep up with flows for all the traffic. Add, both CBWFQ and LLQ are use the same syntax of MQC when configure, this mean it has option for classification, include NBAR. CB and LLQ configuration are similar, different is command for bandwidth and priority. Because two tools are using MQC, so class-map for classification and policy map create group traffic in one gate. CBWFQ and LLQ support 64 queue or classes. Maximum queue can change, with maximum value and default length by router. Queue exist when it don’t be configure. CBWFQ can configure for default group. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 40 - Structure of CBWFQ Figure 2.13 – Operation of class-based WFQ (CBWFQ) - CBWFQ support multi diagram class for classification traffic in correlative queue. - Tail drop is mechanism to default reject CBWFQ. Features of CBWFQ Regulator time of CBWFQ make sure little ratio of bandwidth in every queue. If all queues have big amount of packets, every queue will has a hundred percent bandwidth. However, queues are empty and some queues need bandwidth, it will use bandwidth in little time. Feature CBWFQ Describe Classification Classifies based on anything that MQC commands can match Drop policy Tail drop or WRED, configurable per queue. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 41 - Number of queues 64 Maximum queue length Can change by type of router and memory. Scheduling inside a single queue FIFO on 36 queues; FIFO or WFQ on class-default queue. Scheduling among all queues The result of the scheduler provides a percentage guaranteed bandwidth to each queue. Table 2.2 – Feature WFQ. 2.7.6. Low-Latency Queuing. Low Latency Queuing sounds like the best queuing tool possible, just based on the name. What packet wouldn’t want to experience low latency? As it turns out, for delay (latency) sensitive traffic, LLQ is indeed the queuing tool of choice. LLQ is simple to understand and simple to configure, assuming you already understand CBWFQ. LLQ is not really a separate queuing tool, but rather a simple option of CBWFQ applied to one or more classes. CBWFQ treats these classes as strict-priority queues. In other words, CBWFQ always services packets in these classes if a packet is waiting, just as PQ does for the High queue. LLQ introduces some new lingo that you may find a little tricky. From one perspective, something like PQ has been added to CBWFQ, so you can expect to read or hear phrases that refer to the low-latency queue as “the PQ.” Someone might say, “What did you put in the PQ?” What he really wants to know is what type of packets you classified and placed into the queue in which you enabled the LLQ feature of CBWFQ. In addition, the queue in which LLQ is enabled is sometimes just called “the LLQ.” Therefore, if you use CBWFQ, and use the Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 42 - priority command to enable LLQ in one of the classes, you are really using LLQ, and that one class with the priority command is “the LLQ” or “the PQ.” LLQ search and action similar CBWFQ, different LLQ is permit some of low- latency queues. LLQ similar PQ, LLQ always serve packets in first queue. Sometime, LLQ can use in different way. If policy has little LLQ, this policy is doing LLQ, and queue can call LLQ. Sometime LLQ is similar PQ. When LLQ add into queue with delay time lower than CBWFQ, it can prevent die of queue PQ. In figure 3.7.6, describe operate of algorithm LLQ. Note, although one component run PQ algorithm is show, but this component was control by bandwidth. Class Priority 1 Class Priority n Class 1 Class n BW Policying Class- default BW Policing Y Tail-drop Tail-drop Tail-drop Y n n n n Y Prioryity queue Y Queue 1 Queue n Default queue CBWFQ Sceduler packets Hardware queuing System Figure 2.14 - Describe operate of LLQ LLQ can permit queue configure same with PQ. So have a question “What queue is serving first?” Actually, LLQ put packets from LLQ queues in inside queue. So, Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 43 - packets in priority different queues are serving, after that queues don’t priority are serving. When we control bandwidth of class, we must correct to change different LLQ. Example, we must correct to change voice data and video data, we can put traffic data has low delay, and we must apart from voice data and video data. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 44 - PART 3: QOS OVER MPLS. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 45 - 2.8. Relation about IPP, DSCP and MPLS EXP. 2.8.1. IPP. Figure 2.15 – Architecture of IP header In IP header has type of service field, show service of packet. Example TCP or UDP. And that field used for implement quality of service. Figure 2.16 - Type of Service (byte) From bit 0 to bit 2: these are three precedence bit. And they used for implement quality of service. But three bit is enough for eight levels, too little. So IETF discovery to make bit used for quality of service can increase. In type of service field, from bit 3 to bit 5 are use. And they become DSCP, they have 6 bit for quality of service and increase 64 level, 64 level enough for implement quality of service. IP precedence value Binary value Priority 0 000 Routine Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 46 - Table 2.3- Describe IPP value 2.8.2. DSCP. Figure 2.17- DSCP byte. DSCP is a part wire-open of IP Precedence. And we have two types forwarding in model DSCP: expedited forwarding (EF) and assured forwarding (AF). With EF: lose packet is hard possible, bandwidth make sure, and end-to-end service must through DiffServ domain. With AF: Define services forwarding will make sure, and through a DiffServ domain. DSCP value Binary Name DSCP 0 000000 Default DSCP 8 001000 CS1 DSCP 16 010000 CS2 DSCP 24 011000 CS3 DSCP 32 100000 CS4 DSCP 40 101000 CS5 DSCP 48 110000 CS6 DSCP 56 111000 CS7 DSCP 10 001010 AF11 DSCP 12 001100 AF12 1 001 Priority 2 010 Immediate 3 011 Flash 4 100 Flash override 5 101 Critical 6 110 Internetwork Control 7 111 Network Control Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 47 - DSCP 14 001110 AF13 DSCP 18 010010 AF21 DSCP 20 010100 AF22 DSCP 22 010110 AF23 DSCP 26 011010 AF31 DSCP 28 011100 AF32 DSCP 30 011110 AF33 DSCP 34 100010 AF41 DSCP 36 100100 AF42 DSCP 38 100110 AF43 DSCP 46 101110 EF Table 2.4- DSCP value CS: Class selector EF: Expedited forwarding AF: Assured forwarding. 2.8.3. MPLS EXP. In IP header has ToS field, and ToS has three bit used for implement quality of service in IP network. In MPLS network, MPLS header has three MPLS bit, and three Exp bit are similar Precedence in IP header. And three bit Exp is used for implement quality of service in MPLS network. Figure 2.18 – MPLS header Exp bits in MPLS header are similar Precedence bits in IP header. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 48 - 2.9. DiffServ with IP packets. Refer to Figure 2.19 to refresh your memory about what the IP header looks like. Figure 2.19- IP header field Figure 2.20 shows how the TOS field is divided. Figure 2.20 – The TOS Byte of the IP Header Defining the Precedence Bits The usage of the precedence bits for QoS is now widely used throughout the world for many networks. The drawback of the precedence bits, however, is that only three exist, which means you can have only eight levels of service. Therefore, the IETF decided to dedicate more bits for QoS. The four TOS bits were deprecated, and three of them were assigned to DiffServ QoS, in addition to the three precedence bits. DiffServ ended up with six bits, providing more than enough levels of QoS. Figure 2.21 shows you which bits of the TOS byte are used for DiffServ. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 49 - Figure 2.21-The TOS Byte of the IP Header Defining the DSCP Two types of forwarding classes within the DiffServ model are defined: expedited forwarding (EF) and assured forwarding (AF). EF is a low loss, low latency, low jitter, assured bandwidth, end-toend service through a DiffServ domain. AF defines different services of forwarding assurances through a DiffServ domain. Four classes of AF are defined, each with three drop precedence. AF classes are noted as Afij, with I being 1 to 4 for the class and j being 1 to 3 for the drop precedence. The first three bits of the six-bit DSCP field define the class, the next two bits define the drop precedence, and the last bit is reserved. The higher the drop precedence inside a class, the more likely the packet is to be dropped, relative to the other packets with lower drop precedence when congestion occurs. Four classes exist for the traffic, and three levels exist for drop precedence. AF23, for example, denotes class 2 and drop precedence 3. Table 2.5 - Recommended Values for the Four AF Classes Name DSCP (Binary) DSCP(Decimal) AF11 001010 10 AF12 001100 12 AF13 001110 14 AF21 010010 18 AF22 010100 20 AF23 010110 22 AF31 011010 26 AF32 011100 28 AF33 011110 30 AF41 100010 34 AF42 100100 36 AF43 100110 38 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 50 - Table 2.6- Four AF Classes and Three Drop Precedence 2.10. DiffServ with MPLS packets. In figure 2.18, there are three EXP, or experimental, bits. They are called experimental, but they are really used only for QoS. You can use these bits in the same way that you use the three precedence bits in the IP header. If you use these three bits for QoS, you can call the label switched path (LSP) an E-LSP, indicating that the label switching router (LSR) will use the EXP bits to schedule the packet and decide on the drop precedence. However, when you are using MPLS, you have another option for implementing QoS for the labeled packets. An LSP is a signaled path through the network between two routers. You can use the label on top of the packet to imply part of the QoS for that packet. However, then you need one label per class for each flow of traffic between the two endpoints of the LSP. Therefore, the signaling protocol has to be able to signal a different label for the same LSP or prefix. Such an LSP is called an L-LSP, indicating that the label implicitly holds part of the QoS information. With an L-LSP, the EXP bits still hold part of the QoS, but only the drop precedence, whereas the label indicates the class. With an E-LSP, the EXP bits hold both the class and the drop precedence information. When an LSR forwards a labeled packet, it needs only to look up the top label in its label forwarding table (LFIB) to decide where to forward the packet. The same is true for the QoS treatment. The LSR needs only to look at the EXP bits of the top label to determine how to treat this packet. Remember that QoS constitutes traffic marking, congestion management, congestion avoidance, and traffic conditioning and that you can use low-latency queuing (LLQ), class- Drop precedence Class 1 Class 2 Class 3 Class 4 Low 001010 010010 011010 100010 Medium 001100 010100 011100 100100 High 001110 010110 011110 100110 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 51 - based weighted fair queueing (CBWFQ), weighted random early detection (WRED), policing, and shaping to implement this for IP packets. You can use the same features to implement QoS based on the EXP bits for labeled packets. For example, WRED has been modified to look at the EXP bits to determine the drop precedence of labeled packets when being queued. The preferred way to configure MPLS QoS in Cisco IOS is by means of the Modular Quality of Service Command Line Interface (MQC). MQC is an easy, straightforward way of configuring the different QoS building blocks on the router. Imposition Swap and Imposition Disposition Figure 2.22- Imposition, Disposition and Swap of MPLS labels Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 52 - 2.11. DiffServ Tunneling Modes for MPLS networks. Diffserv Tunneling Modes introduces a new Per−Hop−Behavior (PHB), which allows differentiated QoS in a provider’s network. The tunneling mode is defined at the edge of the network, normally in the PE label switch routers (LSRs) (both ingress and egress). You may need to make changes in the P routers; you must also consider what occurs when the topmost label is removed from a packet due to Penultimate−Hop−Popping (PHP). It may be necessary to copy the MPLS EXP value from the top label that is being popped to the newly exposed label; this does not always apply to all tunneling modes. The MPLS network support of Diffserv specification defines these tunneling modes: Pipe. Short-Pipe. Uniform. The Tunneled Diffserv information is the QoS of the labeled packets or the precedence/DSCP of the IP packets coming into the ingress LSR of the MPLS network. The LSP DiffServ information is the QoS (the value of the EXP bits) of the MPLS packets transported on the LSP from the ingress LSR to the egress LSR. The Tunneled DiffServ information is the QoS information that needs to get across the MPLS network transparently, whereas the LSP DiffServ information is the QoS information that all LSRs in this MPLS network use when forwarding the labeled packet. 2.11.1. Pipe Model. In the Pipe model, the following rules apply: The LSP DiffServ information is not necessarily (but might be) derived from the Tunneled DiffServ information on the ingress LSR. On an intermediate LSR (a P router), the LSP DiffServ information of the outgoing label is derived from the LSP DiffServ information of the incoming label. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 53 - On the egress LSR, the forwarding treatment of the packet is based on the LSP DiffServ information, and the LSP DiffServ information is not propagated to the Tunneled DiffServ information. Figure 2.23 - Pipe model If the MPLS network is receiving IP packets on the ingress LSR and the MPLS network is using E-LSPs only, the Pipe model becomes a bit easier to explain. The Tunneled DiffServ information is the precedence bits or the DSCP of the IP packet. The LSP DiffServ information is the EXP bits value of the labels in the MPLS network. The forwarding treatment (classifying and discard behavior) of IP packets is based on the precedence bits or DSCP in the IP header. This is called the IP PHB (per-hop behavior) hereafter. The forwarding treatment op MPLS packets is based on the EXP bits. This is called the MPLS PHB (per-hop behavior) hereafter. The rules for the Pipe model now translate into the following: The EXP bits can be copied from the IP precedence or set through configuration on the ingress LSR. On a P router, the EXP bits are propagated from incoming label to outgoing label. On the egress LSR, the forwarding treatment of the packet is based on the MPLS PHB (EXP bits), and the EXP bits are not propagated to the IP precedence. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 54 - 2.11.2. Short-Pipe Model. The short-pipe model represents a small variation of the pipe model. It also guarantees that there are no changes to the tunneled PHB marking, even if an LSR re-marks the LSP PHB marking. The short-pipe model shares the same ability of the pipe model to allow an MPLS network to be transparent from the DiffServ point of view. The short-pipe model differs, however, on how the LSP egress infers the packet PHB. The LSP egress uses the tunneled PHB marking to infer the packet PHB and serve the packet consequently. Given this difference with the pipe model, an MPLS network may implement LSPs using the short-pipe model regardless of whether LSRs perform PHP. Figure 2.24 – Short pipe model. The Short Pipe model is similar to the Pipe model, with one difference. The forwarding treatment on the egress LSR is different for the Short Pipe model. Therefore, the third bullet becomes this: Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 55 - On the egress LSR, the forwarding treatment of the packet is based on the Tunneled DiffServ information, and the LSP DiffServ information is not propagated to the Tunneled DiffServ information. If the MPLS network is receiving IP packets on the ingress LSR, that third bullet becomes this: On the egress LSR, the forwarding treatment of the packet is based on the IP PHB (IP precedence), and the EXP bits are not propagated to the IP precedence. 2.11.3. Uniform Model. The Uniform model is quite different from the Pipe or Short Pipe model. In the Uniform model, the following rules apply: The LSP DiffServ information must be derived from the Tunneled DiffServ information on the ingress LSR. On an intermediate LSR (a P router), the LSP DiffServ information of the outgoing label is derived from the LSP DiffServ information of the incoming label. On the egress LSR, the LSP DiffServ information must be propagated to the Tunneled DiffServ information. Notice the change in the first bullet: The LSP DiffServ information must be derived from the Tunneled DiffServ information on the ingress LSR. On the egress LSR, the Tunneled DiffServ information is derived from the LSP DiffServ information. This means that a packet belongs to the same QoS class at any time. The QoS information is always present in the topmost label or in the IP header if the packet is not labeled. The MPLS network does not have an impact on the QoS information, but it does switch the packets through the MPLS network, of course. You can instruct the router to change the EXP bits of the top label(s) through configuration (by using MQC in Cisco IOS) anywhere in the MPLS cloud. This only changes the outer QoS information, or the LSP DiffServ information. This change in the LSP DiffServ information is not Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 56 - propagated down to the Tunneled DiffServ information in the Pipe model and Short Pipe model on the egress LSR. It is, however, propagated on the egress LSR when you are using the Uniform model. Figure 2.25 – Uniform model The uniform model makes the LSP an extension of the DiffServ domain of the encapsulated packet. In this model, a packet only has a single meaningful PHB marking (which resides in the most recent encapsulation). LSRs propagate the packet PHB to the exposed encapsulation when they perform a pop operation. This propagation implies that any packet re-marking is reflected on the packet marking when it leaves the LSP. The LSP becomes an integral part of the DiffServ domain of the packet as opposed to the transparent transport that the pipe and short-pipe models provided. This model proves useful when an MPLS network connects other DiffServ domain and all networks (including the MPLS network) need to behave as a single DiffServ domain. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 57 - 2.12. Steps implement QoS over MPLS. Step 1: Used ACL for classification traffic after we determine traffic in network. Router(config)#access-list 100 permit tcp any any eq 20 Router(config)#access-list 101 permit tcp any any eq 80 Step 2: Create class input for every traffic after we classification. Router(config)#class-map match-any ftpinput Router(config-cmap)#match access-group 100 Router(config)#class-map match-any httpinput Router(config-cmap)#match access-group 101 Step 3: Marking for class. Router(config)#policy-map IN Router(config-pmap)#class ftpinput Router(config-pmap-c)#set ip dscp af11 Step 4: Create class output and match exp value correlative with precedence value or DSCP value. Router(config)#class-map match-any ftpoutput Router(config-cmap)#match mpls experimental topmost 1 Step 5: Create policy bandwidth for every class output Router(config)#policy-map OUTPUT Router(config-pmap)#class ftpoutput Router(config-pmap-c)#bandwidth 10 Step 6: Apply interface Router(config)#interface serial 1/1 Router(config-if)#service-policy input INPUT Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 58 - CHAPTER 3: NETWORK DESIGN AND IMPLEMENT Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 59 - Before design and implement network for Hutech, we introduce about main model of Hutech. In this model, include three branches: -Branch 1 in Binh Thanh district, in this branch includes one router connect to branch 2 in Thu Duc district and branch 3 in Phu Nhuan district, switch for connect LAN network, video device, Web server, Mail server and SQL server. -Branch 2 in Thu Duc district, this branch connect with branch 1, in this branch include one router to connect branch 1, switch for LAN network, video device and SQL server. -Branch 3 similar branch 2, branch 3 in Phu Nhuan district, branch 3 connect to branch 1, include one router, switch for LAN network, video device and SQL server. Three branches connect together, type of connect between routers use Frame- Relay. Because cost for lease-line very high and bandwidth isn’t enough, so obstruct always occur. To troubleshoot this problem, we have solution can make Hutech network better. And solutions we show are QoS over MPLS. Before we don’t apply QoS over MPLS for Hutech network, traffic data always hold bandwidth about 50 percent. So other bandwidth doesn’t enough for traffic video. To settle that problem, we apply technology QoS over MPLS for hutech network. In this model, we carry traffic, both traffic are ftp and video. And result we show detail after. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 60 - 3.1.BUILING SOLUTION FOR HUTECH NETWORK 3.1.1. Real model of Hutech network. Figure 3.1 – Hutech network Explain about connection in Hutech network: Router KTCN 3 connects to Router KTCN 1, type of connect is Frame-Relay technology. Router KTCN 1 connects to Router KTCN 2 , type of connect is Frame-Relay technology. In branch DH KTCN 1, Router KTCN 1 connects to internet, use ADSL technology. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 61 - 3.1.2. Solution model for Hutech network. Figure 3.2- Solution model for Hutech network Explain about connection in solution model: Router KTCN 3 connects to Router KTCN 1, type of connect is MPLS technology. Router KTCN 1 connects to Router KTCN 2 , type of connect is MPLS technology. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 62 - 3.2. Building simulation model to resolve for Hutech network. 3.2.1. Simulation model. Figure 3.3 – Simulation model Device description Three routers, we use Dynamip software to simulation router. Router KTCN 1: -Cisco router 7200. -Ram of router: 96Mb. -IOS router: c7200-jk9o3s-mz.123-18.BIN. Router KTCN 2: -Cisco router 7200. -Ram of router: 96Mb. -IOS router: c7200-jk9o3s-mz.123-18.BIN Router KTCN 3: -Cisco router 7200. -Ram of router: 96Mb. -IOS router: c7200-jk9o3s-mz.123-18.BIN Two PCs: one PC is client, another PC is server. Video device: we use webcam to simulation camera. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 63 - Software of PC Client: Assign IP 172.168.1.2/16, default-gateway: 172.168.1.1 -Window XP. -Install Total Command 7.0 for connect to FTP server. Server: Assign IP 192.168.1.2/24, default-gateway: 192.168.1.1 -Window XP. -Run FTP service. -Run Netflow software for monitor traffic in network. -Install software LEADTOOLSMultimedia for simulation video traffic. Description about connection in simulation model Router KTCN 1: interface Fastethenet 0/0 connect to Server PC, interface S1/0 of router KTCN 1 connect to interface S1/0 of router KTCN 2. -Fastethenet 0/0: assign IP 192.168.1.1/24 -Serial 1/0: assign IP 10.1.1.1/24 Router KTCN 2: interface S1/1 of router KTCN 2 connect to interface S1/0 of router KTCN 3. -Serial 1/0: assign IP 10.1.1.2/24. -Serial 1/1: assign IP 10.2.2.1/24. Router KTCN 3: interface Fastethenet 0/0 of router KTCN 3 connect to client PC. -Fastethenet 0/0: assign IP 172.168.1.1/16. -Serial 1/0: assign IP 10.2.2.2/24. In router KTCN 3 we add command to configure Netflow operation. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 64 - 3.2.2.Implement QoS over MPLS in simulation model. Assign IP address for interfaces in router KTCN3, KTCN2, KTCN1 KTCN3(config)#interface fastethenet0/0 KTCN3(config-if)#ip address 172.168.1.1 255.255.0.0 KTCN3(config)#interface serial1/0 KTCN3(config-if)#ip address 10.2.2.2 255.255.255.0 KTCN2(config)#interface serial1/0 KTCN2(config-if)#ip address 10.2.2.1 255.255.255.0 KTCN2(config)#interface serial1/1 KTCN2(config-if)#ip address 10.1.1.2 255.255.255.0 KTCN1(config)#interface fastethenet0/0 KTCN1(config-if)#ip address 192.168.1.1 255.255.255.0 KTCN1(config)#interface serial1/0 KTCN1(config-if)#ip address 10.1.1.1 255.255.255.0 Active MPLS on router KTCN1, KTCN2, KTCN3 Note: Two routers KTCN1 and KTCN3 have only MPLS on interface serial 1/0, with router KTCN2 both interface serial1/0 and serial1/1 have MPLS. KTCN1(config)#ip cef KTCN1(config)#interface serial1/0 KTCN1(config-if)#mpls ip KTCN1(config-if)#mpls label protocol ldp KTCN1(config-if)#tag-switching ip Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 65 - KTCN2(config)#ip cef KTCN2(config)#interface serial1/0 KTCN2(config-if)#mpls ip KTCN2(config-if)#mpls label protocol ldp KTCN2(config-if)#tag-switching ip KTCN2(config)#interface serial1/1 KTCN2(config-if)#mpls ip KTCN2(config-if)#mpls label protocol ldp KTCN2(config-if)#tag-switching ip KTCN3(config)#ip cef KTCN3(config)#interface serial1/0 KTCN3(config-if)#mpls ip KTCN3(config-if)#mpls label protocol ldp KTCN3(config-if)#tag-switching ip Note: #ip cef //active cisco express forwarding #mpls label protocol ldp //distribution label by ldp #tag-switching ip //active switch ip on MPLS Configure QoS over MPLS Router KTCN1 (router KTCN1 must imposition label and disposition) Classification flow traffic from server to client by access-list KTCN1(config)#access-list 100 petmit tcp any any eq 20 KTCN1(config)#access-list 100 petmit tcp any any eq 21 KTCN1(config)#access-list 101 petmit tcp any any Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 66 - Note: #access-list 100 // used for classification ftp traffic #access-list 101 //used for classification video traffic Create class input for classification traffic. KTCN1(config)#class-map match-any ftp-in KTCN1(config-cmap)#match access-group 100 KTCN1(config-cmap)#match not access-group 101 KTCN1(config-cmap)#match protocol ftp KTCN1(config)#class-map match-anh video-in KTCN1(config-cmap)#match access-group 101 KTCN1(config-cmap)#match not access-group 100 Note: #class-map match-any ftp-in //create class ftp #match access-group 100 // match traffic ftp into class ftp-in #match protocol ftp // match protocol ftp into class ftp-in Marking for every class KTCN1(config)#policy-map IN KTCN1(config-pmap)#class ftp-in KTCN1(config-pmap-c)#set ip dscp AF13 KTCN1(config-pmap)#class video-in KTCN1(config-pmap-c)#set ip dscp CS4 Note: #policy-map IN // create policy #set ip dscp AF13 //marking class ftp-in with dscp AF13, similar with CS4 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 67 - Create class output after we marking KTCN1(config)#class-map match-any mpls-ftp-out KTCN1(config-cmap)#match mpls experimental topmost 1 KTCN1(config)#class-map match-any mpls-video-out KTCN1(config-cmap)#match mpls experimental topmost 4 Note: #match mpls experimental topmost 1// match MPLS EXP value 1 in the topmost label #match mpls experimental topmost 4// match MPLS EXP value 4 in the topmost label Create policy for traffic output KTCN1(config)#policy-map OUT KTCN1(config-pmap)#class mpls-ftp-out KTCN1(config-pmap-c)#bandwidth percent 20 KTCN1(config-pmap)#class mpls-video-out KTCN1(config-pmap-c)#priority percent 70 KTCN1(config-pmap)#class class-default KTCN1(config-pmap-c)#fari-queue Note: #bandwidth percent 20// active CBWFQ and class mpls-ftp-out has 20 percent total bandwidth #priority percent 70// active LLQ and class mpls-video-out has 70 percent total bandwidth #fair-queue //active WFQ Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 68 - Apply into interface KTCN1(config)#interface fastethernet0/0 KTCN1(config-ig)#service-policy input IN KTCN1(config)#interface serial 1/0 KTCN1(config-ig)#service-policy output OUT Router KTCN2 (Forwarding packet) Copy down MPLS EXP value KTCN2(config)#class-map match-any mpls-ftp-in KTCN2(config-cmap)#match mpls experimental topmost 1 KTCN2(config)#class-map match-any mpls-video-in KTCN2(config-cmap)#match mpls experimental topmost 4 KTCN2(config)#policy-map IN KTCN2(config-pmap)#class mpls-ftp-in KTCN2(config-pmap-c)#set qos-group mpls experimental topmost KTCN2(config-pmap)#class mpls-video-in KTCN2(config-pmap-c)#set qos-group mpls experimental topmost #match mpls experimental topmost 1// match MPLS EXP value 1 in the topmost label #set qos-group mpls experimental topmost // Sets a group ID that can be used later to classify packets Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 69 - KTCN2(config)#class-map match-any mpls-ftp-out KTCN2(config-cmap)#match qos-group 1 KTCN2(config)#class-map match-any mpls-video-out KTCN2(config-cmap)#match qos-group 4 KTCN2(config)#policy-map OUT KTCN2(config-pmap)#class mpls-ftp-out KTCN2(config-pmap-c)#bandwidth percent 20 KTCN2(config-pmap)#class mpls-video-out KTCN2(config-pmap-c)#priority percent 70 KTCN2(config-pmap)#class class-default KTCN2(config-pmap-c)#fari-queue Apply interface KTCN2(config)#interface serial1/0 KTCN2(config-if)#service-policy input IN KTCN2(config)#interface serial1/1 KTCN2(config-if)#service-policy output OUT Router KTCN3 (Copy down MPLS EXP to IP Precedence or DSCP) Copy down MPLS EXP value KTCN3(config)#class-map match-any mpls-ftp-in KTCN3(config-cmap)#match mpls experimental topmost 1 KTCN3(config)#class-map match-any mpls-video-in KTCN3(config-cmap)#match mpls experimental topmost 4 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 70 - KTCN3(config)#policy-map IN KTCN3(config-pmap)#class mpls-ftp-in KTCN3(config-pmap-c)#set qos-group mpls experimental topmost KTCN3(config-pmap)#class mpls-video-in KTCN3(config-pmap-c)#set qos-group mpls experimental topmost #match mpls experimental topmost 1// match MPLS EXP value 1 in the topmost label #set qos-group mpls experimental topmost // Sets a group ID that can be used later to classify packets KTCN3(config)#class-map match-any mpls-ftp-out KTCN3(config-cmap)#match qos-group 1 KTCN3(config)#class-map match-any mpls-video-out KTCN3(config-cmap)#match qos-group 4 KTCN3(config)#policy-map OUT KTCN3(config-pmap)#class mpls-ftp-out KTCN3(config-pmap-c)#bandwidth percent 20 KTCN3(config-pmap)#class mpls-video-out KTCN3(config-pmap-c)#priority percent 70 KTCN3(config-pmap)#class class-default KTCN3(config-pmap-c)#fari-queue Apply interface KTCN3(config)#interface serial1/0 KTCN3(config-if)#service-policy input IN KTCN3(config)#interface fastethernet0/0 KTCN3(config-if)#service-policy output OUT Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 71 - Configure Netflow operate in router KTCN3 KTCN3(config)#interface fastethernet0/0 KTCN3(config-if)#ip route-cache flow KTCN3(config)#ip flow-export destination 192.168.1.222 9996 KTCN3(config)#ip flow-export source fastethernet0/0 KTCN3(config)#ip flow-export version 5 KTCN3(config)#ip flow-cache timeout active 1 KTCN3(config)#ip flow-cache timeout inactive 15 KTCN3(config)#snmp-server communication ktcn3 Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 72 - 3.3. Get Result Compare bandwidth before and after we apply QoS over MPLS Bandwidth of network before apply QoS over MPLS Figure 3.4 – Get result before implement QoS In figure 3.4, we can see ftp-data traffic more than TCP_App (video traffic) Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 73 - Bandwidth of network after apply QoS over MPLS Figure 3.5 – Get result after implement QoS In figure 3.5, we can see video traffic (TCP_App) more than tcp-app traffic Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 74 - GET RESULT AND DEFINE OF DEVELOP IN SUBJECT GET RESULT: After we implement QoS over MPLS, we were successful and we get result: -Controlling traffic in network, to correct the traffic important example voice traffic, video traffic… -Avoiding obstructed, preference with important traffic. -This model can apply to Hutech network and real network. DEFINE OF DEVELOP IN SUBJECT: Present, IPv6 is implementing in some country, example USA. If we deploy MPLS on IPv6, we can create something new. In IPv6 environment can better security then IPv4, and number of address IPv6 more than address IPv4. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 75 - References [1] www.NetAP.net Cisco Press 2000 Cisco Press MPLS and VPN Architectures [2] www.NetAP.net Cisco Press 2001 Cisco Press Advanced MPLS Design and Implementation [3] www.NetAP.net Cisco Press 2006 MPLS Fundamentals [4] www.NetAP.net Cisco Press 2006 QoS for IPMPLS Networks [5] www[1].NetAP.net Cisco Press 2005 MPLS Configuration on Cisco IOS Software [6] Cisco Press-DQOS.Exam.Certification.Guide [7] www[1].NetAP.net Cisco Press 2004 End-to-End QoS Network Design [8]rfc 3031 Web site: www.cisco.com www.vnpro.org Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 76 - Index [1]Figure 1.1 - Network diagram of Hutech network [2]Figure 1.2 – Solution model of Hutech network [3]Figure 2.1 – MPLS label architecture. [4]Figure 2.2 – Label of Stack. [5]Figure 2.3 – Imposition and disposition. [6]Figure 2.4 – Label Switch Path. [7]Figure 2.5 – IntServ model. [8]Figure 2.6 – DiffServ model. [9]Figure 2.7 – FIFO Queue. [10]Figure 2.8 – Priority Queue. [11]Figure 2.9 – Classification and move packet into SNA Queue. [12]Figure 2.10 – Model of operation WFQ. [13]Figure 2.11 – Describe calculator SN. [14]Figure 2.12 – Describe process of WFQ. [15]Figure 2.13 – Operation of Class-Based WFQ. [16]Figure 2.14 – Describe operate of LLQ. [17]Figure 2.15 – Architecture of IP header. [18]Figure 2.16 – Type of Service. [19]Figure 2.17 – DSCP byte. [20]Figure 2.18 – MPLS header. [21]Figure 2.19 – IP header field. [22]Figure 2.20 – The ToS byte of the IP header define the precedence bits. [23]Figure 2.21 – The ToS byte of the IP header defining the DSCP. [24]Figure 2.22 – Imposition, Disposition and Swap of MPLS labes. Supervisor: Nguyễn Đức Quang QoS over MPLS for Hutech network Student: Trần Quang Hải Đăng - 77 - [25]Figure 2.23 – Pipe Model. [25]Figure 2.24 – Short-Pipe model. [26]Figure 2.25 – Uniform model. [27]Figure 3.1 – Hutech network. [28]Figure 3.2 – Solution model for Hutech network. [29]Figure 3.3 – Simulation model. [30]Figure 3.4 – Get result before implement QoS. [31]Figure 3.5 – Get result after implement QoS. [32]Table 2.1 – Compare IntServ model and DiffServ model. [33]Table 2.2 – Feature WFQ. [34]Table 2.3 – Describe IPP value. [35]Table 2.4 – DSCP value. [36]Table 2.5 – Recommended values for the four AF classes. [37]Table 2.6 – Four AF classes and three drop precedence.