WO2006034644A1 - Reseau en anneau et methode de mise en oeuvre de ses services - Google Patents

Reseau en anneau et methode de mise en oeuvre de ses services Download PDF

Info

Publication number
WO2006034644A1
WO2006034644A1 PCT/CN2005/001580 CN2005001580W WO2006034644A1 WO 2006034644 A1 WO2006034644 A1 WO 2006034644A1 CN 2005001580 W CN2005001580 W CN 2005001580W WO 2006034644 A1 WO2006034644 A1 WO 2006034644A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
ring network
ring
network
node
Prior art date
Application number
PCT/CN2005/001580
Other languages
English (en)
French (fr)
Inventor
Yuxiang Wang
Huaixue Wan
Li Zeng
Original Assignee
Huawei Technologies Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co., Ltd. filed Critical Huawei Technologies Co., Ltd.
Priority to US11/579,029 priority Critical patent/US20070280251A1/en
Priority to EP05792149A priority patent/EP1796328A4/en
Publication of WO2006034644A1 publication Critical patent/WO2006034644A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/28Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/42Loop networks
    • H04L12/437Ring fault isolation or reconfiguration
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/28Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46Interconnection of networks
    • H04L12/4637Interconnected ring systems
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/50Routing or path finding of packets in data switching networks using label swapping, e.g. multi-protocol label switch [MPLS]

Definitions

  • the present invention relates to the field of network communication technologies, and in particular, to a ring network and a service implementation method thereof.
  • MPLS Multi-Protocol Label Switching
  • VPI/VCI Virtual Channel Identification/Virtual Circuit Identification
  • ATM Asynchronous Transfer Mode
  • MPLS technology has gradually become a data network solution with its excellent performance in multi-service/protocol support, MPLS L3/L2 VPN (virtual private network) and traffic engineering.
  • MPLS L3/L2 VPN virtual private network
  • Mainstream technologies and further expand from the core layer of the network to the aggregation layer and access layer.
  • the key to the MPLS protocol is the introduction of the concept of a label. It is a short, easy-to-handle information content that does not contain topological information and only has local meaning. Label is short for easy handling and can usually be referenced directly by index. Only local meaning is for ease of distribution.
  • the basic features of the MPLS technology are: For each packet entering the MPLS domain, the edge router assigns a label to the packet according to the traffic classification of the packet. Since then, all the MPLS routers receive the marked packet directly. Forward by tag, and change the tag to the new tag that you have agreed with the next router. Then, after the tagged message leaves the MPLS domain, it will re-export the IP packet. When such a marked message leaves the MPLS domain, the label is separated by the edge router.
  • the service type of the MPLS packet can be determined by the MPLS edge router according to various parameters of the IP packet, such as the source address, destination address, port number, and service type (TOS) value of the IP.
  • TOS service type
  • TOS service type
  • TOS service type
  • MPLS can establish new forwarding routes in real time to share its traffic to alleviate network congestion.
  • the current mainstream MPLS encapsulation in the industry is Martini encapsulation, which is defined in the I ⁇ TF draft draft-martini-12circuit-encap-mpls-Q4.
  • the Martini encapsulated MPLS frame consists of two layers of labels, the outer layer is a tunnel label, and the inner layer is a VC label.
  • Tag stacks can be nested infinitely, providing unlimited business support.
  • the tunnel LSP1 is a pipe LSP between R1 and R5, and the label switching path is 10 (R1/R6) ⁇ 20 (R6/R5).
  • R6 only completes tunnel label switching; virtual container label LSP2 It is a service LSP between R1 and R5.
  • the label switching path is 10.01 ( R1/R6 ) ⁇ 20.01 ( R6/R5 ). R6 only performs tunnel label switching, and the inner label is invisible to R6.
  • MPLS LSPs usually use 1+1 or 1:1 protection technology.
  • LSP protection is an end-to-end protection. Although this kind of protection can effectively protect the business, there are certain disadvantages:
  • the mesh connection of the working LSP and the mesh connection of the protection LSP, the working LSP and the protection LSP all have the problem of N-square, which causes the network to configure and maintain a large amount of LSP information when there are many network nodes.
  • the protection is implemented by using the MPLS OAM function, which is passive protection, that is, the terminal end of the service is continuously checked. If the OAM frame, such as CV or FDI, finds that the CV is lost or receives the FDI error multiple times, the LSP is considered invalid. Considering that the OAM is in-band transmission, the LSP bandwidth is occupied. Therefore, the OAM frame cannot be sent in real time, and the transmission frequency is not easy. Too fast, resulting in faults not being timely enough to meet carrier-class protection switching requirements;
  • C) 1+1 protection is a dual-issue selection method, resulting in low network bandwidth utilization.
  • the service can get better QoS (quality of service) and protection.
  • RPR Resilient Packet Data Loop
  • RPR adopts double-ring structure: There are two paths between each pair of nodes to ensure high availability. Space reuse mechanism is adopted for loop bandwidth. Unicast data transmission can be performed simultaneously in different parts of the ring, which improves the utilization of loop bandwidth. rate.
  • the RPR ring network can support fast protection of 50 ms.
  • the RPR ring network can adopt two protection mechanisms. One is the source routing mode (Steering), that is, switching directly at the source of the service to ensure the best path for the service. It is a method of Wrapping at two nodes where a failure occurs, similar to the 2-fiber MS-SPRing (Multiplex Segment Shared Protection Ring) of SDH (Synchronous Air Word System).
  • the RPR technology is a MAC (Media Access Control Layer) layer protocol that optimizes data traffic transmission on a ring structure. It can adapt to multiple physical layers and efficiently transmit data of various service types such as voice and image. It combines the economics, flexibility, and scalability of Ethernet technology. It also absorbs the advantages of 50ms fast protection of SDH ring network, and has automatic topology discovery, loop bandwidth sharing, fair distribution, and strict COS (Business Classification) and other functions.
  • MAC Media Access Control Layer
  • the RPR technology also has its limitations. Because the IEEE 802.17 specification is an RPR MAC layer technology designed for a single physical ring or a logical ring (a virtual container VC channel that spans multiple SDH physical rings), the RPR The application is limited to a single ring. It must be terminated when it crosses the ring. That is, the end-to-end bandwidth sharing, fairness mechanism, QoS (quality of service), and protection functions of the cross-ring service cannot be realized.
  • the ⁇ : method is to introduce a Layer 2 or Layer 3 switch between the ingress and egress rings and between multiple rings. This makes the network more complex and the network structure is not clear.
  • Another solution is to use MPLS over RPR (elastic packet data ring bearer multi-protocol label switching) to compensate for the shortcomings of RPR, but this process introduces two layers: RPR layer and MPLS layer, as shown in Figure 2, The complexity of the service processing is increased, and the processing efficiency is low.
  • the frame format of the RPR ring network is a dedicated format, the introduction of the MPLS layer increases the overhead of each data packet, so that the bandwidth utilization rate is reduced;
  • the RPR protection ring can be built based on physical links or sub-channels. As a result, the networking is not flexible enough to provide services, and it is not suitable to build a large network.
  • the shortcomings of the RPR ring network solve the interworking of cross-ring services, provide better QoS protection and rich SLA (Service Level Agreement).
  • a ring network comprising: a plurality of nodes on the same logical layer, used to access a service The ring network or the service is received from the ring network; each phase 4 (the five nodes are connected by two virtual channels in opposite directions, the virtual channel is used to carry suitable service data; Adapting to the business data of the virtual channel.
  • the virtual channel is a label switched path based on multi-coordination and protocol label switching.
  • the virtual channel can also be an asynchronous transfer mode virtual channel connection.
  • the node adapts the carried service data adapted to the virtual 4 channel to the physical chain through an Ethernet media access control protocol or a general framing procedure or a link access procedure or a high speed digital link control connection In the road.
  • the physical link includes: an Ethernet link, a synchronous digital network link, an optical transport network link, and a virtual concatenation group.
  • the physical link is located in a ring network and/or in an intersecting bad and/or mesh network.
  • the services include: Internet service, network service, and asynchronous transmission mode service.
  • the encapsulation format of the service data accessed or received by the node is a standard frame format of multi-protocol label exchange.
  • a service implementation method of a ring network where the ring network includes: a plurality of nodes on the same logical layer, and two virtual channels connected in opposite directions to each adjacent node and used to carry the adaptation to the virtual channel
  • the physical link of the business data the methods include:
  • the service source node obtains a service sink node according to the label switching path scheduling table.
  • the service node performs the service on the ring network.
  • the method further includes: applying ring switching protection and/or source routing protection to the ring network.
  • the step A further includes: binding the virtual channel and the corresponding ring network.
  • the step B includes: statically configuring and maintaining: and dynamically configuring and maintaining the service label exchange path schedule.
  • the dynamically configuring and maintaining the service label switching path scheduling table includes: The protocol and resource reservation protocol configure and maintain the label switching path schedule.
  • the label switching path scheduling table includes: label action information destination port information.
  • the step C further includes: completing the multi-protocol label switching encapsulation of the non-multi-protocol label switching service according to a predetermined rule.
  • the predetermined rules include:
  • the non-multi-protocol label switching service group is divided into different forwarding equivalence classes according to the destination address, and the appropriate label is inserted into the splitting header according to the forwarding equivalence class to which the packet belongs, and the encapsulation of the multi-protocol label exchange is completed;
  • the non-multi-protocol label switching service group is divided into different forwarding equivalence classes, and the appropriate label is inserted into the packet header according to the forwarding equivalence class to which the packet belongs, and the encapsulation of the multi-protocol label switching is completed.
  • the predetermined scheduling algorithm includes: a strict priority scheduling method.
  • the method further includes: adopting a service label switching path for the intra-ring service and the cross-ring service
  • the method further includes:
  • the information of the flow classification includes: a priority domain of the virtual local area network service and/or a service type domain of the Internet Protocol service and/or a priority specified by the administrator.
  • the method further includes:
  • the media access control layer of each node on the ring network measures the utilization of the link connected to the node, and notifies the observed network to the node on the ring network;
  • Each node on the ring network adjusts the rate at which the local node sends data to the ring network according to the fair algorithm protocol and the obtained notification.
  • the present invention is based on MPLS technology, so that The ring network not only has all the functions of the RPR ring network, but also has the following advantages over the RPR ring network: It does not depend on a specific physical layer technology, and can construct the ring network of the present invention in a network implemented by multiple physical layer technologies;
  • the network topology is built to virtual/physical or logical MPLS network, which is more flexible and fast, and can effectively improve the service opening speed, especially the VPN (Proprietary Private Network) service;
  • the virtual MPLS network has one less level than the RPR. Level, so the processing is more simple, the configuration of the service is simpler and clearer, and the protection speed and efficiency of the service are improved.
  • the network adopts the standard MPLS frame format, so that the service does not depend on the ring, and no other 4t assistance technology is needed.
  • the cross-ring end-to-end service provision and the multi-ring intersection/cutting service interworking can be implemented on the ring network of the present invention, and the network bandwidth utilization rate is improved.
  • Different service LSPs on the ring network of the present invention can define different QoS.
  • the parameters and supported SLAs are more abundant. By scheduling the LSP granules according to the QoS parameters negotiated in advance, it is better to guarantee and support the difference.
  • the bearer mode of the service on the ring network is LSP, it is a connection-oriented technology, which can fully utilize the OAM (Operation and Maintenance) function in MPLS: including LSP connectivity check (CV) and LSP fast failure detection. (FFD), Forward Defect Indication (FDI), Backward Defect Indication (BDI), etc., for effective detection and maintenance of services.
  • OAM Operaation and Maintenance
  • FIG. 1 is a schematic diagram of an MPLS network
  • FIG. 2 is a schematic diagram of a hierarchical structure of an RPR ring network in the prior art
  • Figure 3 is a top view of the ring network of the present invention.
  • FIG. 4 is a schematic diagram showing the hierarchical structure of the ring network of the present invention.
  • FIG. 5 is a schematic diagram of a ring network of FIG. 4 with an MPLS LSP as a virtual channel multiplexing layer;
  • FIG. 6 is a schematic diagram of different physical networking modes of the ring network of the present invention.
  • FIG. 7 is a simplified structure of a virtual MPLS ring network formed by configuring a LSP in the network shown in FIG. 6.
  • FIG. 8 is a network topology diagram when SDH/SONET is used as a bearer layer of the ring network of the present invention;
  • FIG. 9 is a ring network shown in FIG. Schematic diagram of the layer structure;
  • FIG. 10 is a network topology diagram when Ethernet is used as a bearer layer of the ring network of the present invention
  • FIG. 11 is a schematic diagram showing a layer structure of the ring network shown in FIG.
  • 12 is a flowchart of a method for implementing a service of a ring network according to the present invention
  • 13 is a schematic diagram of implementation of a ring network service in the present invention
  • 15 is a schematic diagram of protection implementation of a cross-ring service according to the present invention.
  • FIG. 16 is a process for implementing network-based protection by using a ring switching manner in the present invention
  • FIG. 17 is a process for implementing network-based protection by using a source routing method in the present invention.
  • the core of the present invention is to construct a ring network technology that does not depend on a physical topology structure and a specific physical layer technology by using MPLS technology.
  • a virtual MPLS ring network is composed of multiple nodes at the same logical level to form a two-channel bidirectional ring. Type network.
  • the virtual ring network topology is shown in Figure 3, including:
  • each adjacent node is connected by two virtual channels in opposite directions, and is used to transmit service data between nodes; Service data can be carried over different physical links.
  • the two virtual channels form two opposite rings: the west ring and the east ring, the west ring transmits the clockwise traffic, the east ring transmits the traffic counterclockwise, and the ring node accesses or receives the service using the standard MPLS frame structure. There is no need to redefine the frame format.
  • the virtual channel carries the service data of different nodes in the manner of LSP (Label Switched Path). Each node identifies whether it is a local service through the MPLS label of the service. If it is a local service, it strips the service, otherwise it is transmitted to the next adjacent site through the virtual channel.
  • LSP Label Switched Path
  • the virtual channel can be built using a variety of technologies, such as through a label switched path, or through other VPN (Virtual Private Network) leased line technologies, such as ATM VPC (Virtual Path Connection).
  • VPN Virtual Private Network
  • ATM VPC Virtual Path Connection
  • the virtual ring network layer structure is shown in Figure 4:
  • the MPLS service aggregation layer uses MPLS technology to implement service access, transmission, aggregation, and multiplexing through service LSPs.
  • the virtual channel multiplexing layer can be used but not limited to MPLS technology.
  • the LSP is used as a virtual channel for constructing the virtual MPLS ring network of the present invention, or when it is used as the MPLS ring network bearer layer, the network layer structure is as shown in FIG. 5.
  • the virtual channel layer can be omitted.
  • the bearer layer of the virtual MPLS ring network becomes the data link layer, so that the virtual MPLS ring network evolves into an MPLS ring network based on physical links or logical sub-channels, such as GE-based. (gigabit Ethernet) MPLS ring network connected, or MPLS ring network based on VCG (virtual concatenation group) connection.
  • VCG virtual concatenation group
  • the service layer can access a variety of different services, such as IP (Internet) services, ETH (Ethernet) services, ATM (Asynchronous Transfer Mode) services, and so on.
  • IP Internet
  • ETH Ethernet
  • ATM Asynchronous Transfer Mode
  • the MPLS service layer uses MPLS technology to aggregate and reuse services, using standard
  • the MPLS frame format ensures interworking between existing MPLS networks and cross-ring services.
  • the data link layer can be, but is not limited to, data link layer technologies such as Ethernet MAC (Media Access Control), GFP (General Frame Protocol), LAPS (Link Access Procedure), HDLC (Advanced Data Link Control), and the like. That is, the nodes on the ring network can adapt the service data transmitted on the virtual channel to the physical link through Ethernet MAC protocol, GFP, LAPS, HDLC, and the like.
  • the ring network of the present invention has no dependency on the data link layer.
  • the ring network has no dependence on the physical layer, and different physical layer technologies can be used as needed.
  • Ethernet technology Packet over SDH
  • EOS Ethernet over SDH
  • VCAT virtual concatenation
  • POS is an IP packet that encapsulates IP packets by using PPP (Point-to-Point Protocol) and maps to SDH/SONET (Synchronous Digital Hierarchy/Synchronous Optical Network) frames using HDLC frame format, according to a corresponding line. Continuous transmission, which preserves the non-connected nature of IP.
  • PPP Point-to-Point Protocol
  • SDH/SONET Serial Digital Hierarchy/Synchronous Optical Network
  • HDLC frame format is responsible for delimiting the IP data frames encapsulated by PPP on the transmission link.
  • the PPP protocol cuts IP packets into PPP frames to meet the requirements mapped to the SDH/SONET frame structure.
  • EOS mainly defines a mapping method for encapsulating Ethernet frames and then mapping them to VCs (virtual containers) of SDH/SONET. It is located between the Ethernet MAC layer and the SDH of the physical layer as a data link adaptation layer.
  • the main package mapping methods are PPP/HDLC, LAPS and GFP.
  • a traditional MPLS LSP is an end-to-end "connection" or "virtual channel”
  • an MPLS LSP is used as a virtual channel for connecting nodes on the ring network, and a certain bandwidth is allocated for these channels, that is, Formed a virtual MPLS ring network structure, which is the hair Ming's ring network.
  • This method is independent of the physical networking form and can span the limits of the physical network.
  • the physical networking can be a ring network, an intersecting ring, a mesh (mesh) network, and so on.
  • the virtual MPLS ring network in the figure can be simplified into the structure shown in FIG. 7. Although the three network elements R6, R7 and R8 are not physically adjacent, they can be configured into a virtual channel through the channels LSP1 LSP6. MPLS Ring, LSP1, LSP3, and LSP5 form a westbound ring, and transmit traffic clockwise. LSP2, LSP4, and LSP6 form an eastbound ring and transmit traffic counterclockwise. The service between the three NEs is carried in the channel LSP through the service LSP.
  • the virtual channel multiplexing layer of the ring network of the present invention is not required.
  • the bearer layer of the MPLS ring becomes a data link layer, and the channel can be a physical link or a sub-channel, and an MPLS ring channel. Bandwidth is no longer multiplexed.
  • a dual-channel bidirectional network network can be constructed by using VCG as a channel.
  • Each adjacent node is connected by two opposite VCG channels, and two VCG rings are formed in opposite directions. It is called the westbound ring and the eastbound ring, the westbound ring transmits the clockwise traffic, the eastbound ring transmits the traffic counterclockwise, and the traffic transmitted on the ring carries the MPLS label switching path.
  • VCG1, VCG3, VCG5, and VCG7 form a westbound ring, which transmits traffic clockwise.
  • VCG2, VCG4, VCG6, and VCG8 form an eastbound ring and transmit traffic counterclockwise.
  • the ring network structure of this technology is shown in Figure 9. Where SDH and virtual concatenation form a virtual The physical layer of the MPLS ring network, the data link layer uses GFP or LAPS/HDLC to form the bearer layer of the MPLS ring network.
  • Ethernet can also be directly used as the bearer layer of the virtual MPLS ring network.
  • the GE port of the router is used to connect the nodes into a dual-fiber bidirectional ring structure, the westbound ring transmits traffic clockwise, and the eastbound ring transmits traffic counterclockwise.
  • the ring network structure of this technology is shown in Figure 11.
  • the MAC layer of the Ethernet constitutes the physical layer of the virtual MPLS ring network, and the data link layer uses GFP or LAPS HDLC to form the bearer layer of the MPLS ring network.
  • the virtual channel is configured on each node to form an east and west bidirectional ring network, that is, a virtual channel is constructed on the same logical layer, and a certain bandwidth is allocated to form an east ring and a west ring.
  • Virtual channels can be constructed using a variety of technologies. For example, MPLS LSPs are used to build virtual channels between the MPLS service aggregation layer and the data link layer. Physical or logical subchannels can also be built at the physical layer based on physical links or logical subchannels.
  • the virtual MPLS ring network is evolved into an MPLS ring network based on a physical link or a logical sub-channel, such as an MPLS ring network based on a GE connection or an MPLS ring network based on a VCG connection.
  • Step S12 Bind the virtual channel and the corresponding ring network.
  • multiple virtual MPLS rings may be constructed, and each port may also contain multiple virtual channels. Therefore, each virtual channel to be built needs to be bound to its corresponding ring network. This consists of a number of adjacent or non-adjacent ports in the physical network forming a complete loop through the constructed virtual channels.
  • the different services accessed by the service layer are first adapted to the virtual channel multiplexing layer through the MPLS encapsulation, and then adapted to the physical layer through the encapsulation of the data link layer.
  • Step S13 Establish a service label switching path scheduling table at each node.
  • the label switching path scheduling table includes information such as a label action and a destination port.
  • the label action includes a pop-up label, a label insertion, and a label exchange.
  • the table may be established in a static configuration manner or by using an LDP (Label Distribution Protocol) protocol.
  • the RSVP Resource Reservation Protocol
  • the RSVP Resource Reservation Protocol
  • Step S14 At the service source node entry, complete the multi-label protocol exchange encapsulation of the non-multi-label protocol exchange service according to a predetermined rule.
  • the detailed process can refer to Figure 13, which includes four nodes. A, B, C, D.
  • the non-MPLS service packet is divided into different forwarding equivalence classes according to a predetermined rule, and the appropriate label is inserted into the packet header according to the forwarding equivalence class to which the packet belongs, that is, the completion is completed. Encapsulation of protocol label switching.
  • the predetermined rule is specified by the user, for example, the non-MPLS service can be grouped according to the destination address, QOS (Quality of Service) requirements, and the like. If it is an MPLS service, this step is not required.
  • Step S15 The service source node obtains the service sink node according to the service label switching path scheduling table. Specifically, according to the service label switching path scheduling table, after performing a label action (popping a label, adding a label, and exchanging a label), the outbound port is found, that is, the service sink node is obtained.
  • Step S16 The service is scheduled on the ring network according to a predetermined scheduling algorithm.
  • the user service can be scheduled on the ring network according to the strict priority scheduling algorithm to provide users with better level service.
  • Step S17 The services of different nodes are multiplexed in the same virtual channel by the label switching path for transmission.
  • the services multiplexed in the virtual channel can be carried over different physical links.
  • the aforementioned VCAT of SDH/SONET the multiplexing service is mapped to the SDH/SQNET VC through the GFP/LAPS/HDLC package (virtual container). ); or directly use Ethernet as the physical link carrying the multiplexing service, and transmit it through the GE port of the router.
  • Step S18 Perform label switching through the intermediate node to transmit the user service to the service sink node. For example, in Figure 13, after the label exchange of the intermediate node B, it is quickly forwarded to the next node ⁇ 3.
  • Step S19 The service sinks the service to the ring network.
  • the node is a sink node of the label switching path, and the user service is scheduled to perform the ring network, and the client layer signal is further processed.
  • the ring network of the present invention uses the standard MPLS frame format to transmit user services, so that the service does not depend on the ring. Therefore, devices outside the ring network can identify the user frame of the ring network without any processing, thereby conveniently Realize cross-ring end-to-end service provision and multi-ring intersection/tangential service interworking.
  • FIG. 14 is a schematic diagram of the implementation of the cross-ring service on the ring network of the present invention:
  • ring network A includes four nodes: node A,
  • Ring B consists of four nodes: nodes E, F, G, H.
  • 411 is the actual path of the service on the ring network A
  • 412 is the actual path of the service between the rings
  • 413 is the service The actual path on ring B.
  • the paths 411, 412, and 413 respectively correspond to one LSP, that is, on the ring network A, from the node D to the node A, which is an LSP: LSP1; on the ring network B, from the node, ⁇ E to the node H, It is an LSP: LSP3.
  • LSP label switching from LSP1 to LSP2 and label switching from LSP2 to LSP3 are completed.
  • the cross-ring service can use MPLS LSP scheduling to implement service scheduling between the ring network and the ring network.
  • the cross-ring service scheduling can be completed without other conversion.
  • the present invention also adopts the following various protection measures for the ring network:
  • MPLS OAM Operaation and Maintenance
  • the implementation is as follows:
  • the working LSP and the backup LSP are reversely configured.
  • the working LSP is configured in the westbound ring and the protection LSP is configured in the eastbound ring.
  • the MPLS OAM function is used to detect the situation. Then enable the alternate LSP.
  • MPLS OAM Protection of cross-ring services: MPLS OAM is used to implement 1:1 or 1+1 protection of LSPs.
  • the implementation is as follows: The working LSP and the standby LSP cross different links of the intersecting ring. After the working LSP is faulty, the MPLS OAM function is used to monitor the situation and enable the standby LSP.
  • the MPLS 0AM (Operation and Maintenance) is mainly guaranteed by the MPLS OAM frame defined by ITU-T Rec. Y. 1711.
  • the currently defined frame types are CV (Connectivity Check), FDI (Forward Defect Indication), BDI ( Backward defect indication), performance message, loopback request, loopback response, but only CV, FDI, BDI are defined in the specific format and operating procedures.
  • (a) Connectivity check The CV stream is generated at the source LSR of the LSP and sent at the speed of 1/S. The LSR is terminated at the LSP. The CV packet carries the network unique identifier (TTSI), which is able to detect all The basis of the defect.
  • TTSI network unique identifier
  • 3 ⁇ 4 indication The BDI stream is inserted on the return path (such as a returning LSP) to notify the upstream LSR of the defect detected in the LSR (Label Switching Router) sink of the downstream LSP. (source point of forward LSP).
  • the protection of the traditional ring network generally has two kinds of wrapping (ring switching protection) and s teer ing (source routing protection).
  • the virtual MPLS ring of the present invention can be divided into a unidirectional protection ring and a bidirectional protection ring according to the manner in which the nodes access the service.
  • the former specifies one of the two opposite rings to be used as a working channel for transmitting under normal conditions.
  • the upper and lower sides of the service are always completed on the working channel, and the other ring is used as a protection channel for protection against network failure.
  • the working channel; the latter can transmit services in both directions at the same time. Under normal circumstances, the service takes a short path, and after the failure occurs, the service is transmitted through the long path.
  • the protection of the MPLS ring can be protected by Wrapping and s teering. The following two methods of protection are described separately.
  • Wraping protection can be performed without the help of signaling. It only needs to detect the channel link status of the service layer. When the service layer has SF (signal failure), the channel between the working channel and the protection channel is corresponding to the channel on the side of the fault point. .
  • SF signal failure
  • the topology information is exchanged between the nodes of the ring network through signaling, and each node knows the state of the network (optional);
  • Figure 16 shows the implementation of unidirectional protection ring switching with LSP as the bearer layer:
  • LSP1, LSP3, LSP5, LSP7, LSP9 and LSP11 form the westbound ring of the MPLS Ring, and serve as the working channel.
  • LSP2, LSP4, LSP6, LSP8, LSP10, and LSP12 form the eastbound ring of the MPLS Ring and serve as a protection channel to transmit traffic counterclockwise.
  • LSPQ is a service from R1 to R4.
  • JE is often clockwise through R2 and R3. When it is transmitted to the R4 site, it is stripped by the destination station R4.
  • both R2 and R3 are connected to the working channel and the protection channel on the side adjacent to the fault point.
  • the LSPQ should be sent to R3 in the R2 node encapsulated ll channel LSP3.
  • the fault After the fault occurs, it is encapsulated into the protection channel LSP2 and sent to R1 counterclockwise, and reaches the R3 through the protection channel of R6, R5 and R4.
  • R3 switches it to the working channel and encapsulates it in channel LSP5, and sends it clockwise to R4.
  • R4 finds it is a local service and then strips it.
  • the protection channel can transmit additional services when the network is normal.
  • the s teer ing protection requires the assistance of signaling. Generally, under normal circumstances, the service takes a short path. After the network fails, the service is recalculated through topology discovery.
  • the topology information is exchanged between the nodes of the ring network through signaling, and each node knows the state of the network;
  • the source node of the LSP After receiving the information, the source node of the LSP immediately redirects the service LSP to another virtual channel ring to implement source route protection.
  • Figure 17 shows the source route protection implementation of the bidirectional protection ring with LSP 7 carrier layer: LSP1, LSP3, LSP5, LSP 7, LSP9 and LSP11 form the westbound ring of the MPLS Ring, and transmit traffic clockwise.
  • LSP2, LSP4, LSP6, LSP8, LSP10, and LSP12 form the eastbound ring of the MPLS Ring and transmit traffic counterclockwise.
  • LSPQ and LSPN are the two industries between R1 and R3 nodes. 0
  • LSPQ passes clockwise through R2 to R3 and is stripped by R3.
  • LSPQ passes through R2 to R1 counterclockwise and is stripped by R1.
  • R1 discovers that the LSPQ is unreachable, reverses the LSPQ to the eastbound ring, and reverses the traffic to R3 through R6, R5, and R4, and is stripped by R3.
  • R3 finds that the LSPN is unsuccessful in clockwise direction. The LSPN is switched over to the westbound ring.
  • the traffic is clocked through R4, R5, and R6, and is transmitted to R1 and stripped by R1.
  • a certain protection bandwidth should be reserved on both rings to protect the working service in case of failure. The reserved bandwidth can transmit additional services when the network is normal.
  • the Wrapping protection mode has a fast switching speed and does not require topology information. However, when the switching occurs, the service is entangled in the network, resulting in low bandwidth utilization.
  • the Steer ing protection method has high bandwidth utilization, but requires network topology information. When the network is faulty, it needs to wait for a re-topology discovery before it can find the best route, so the switching time is longer.
  • the Wrapping + Steering method provides a better solution.
  • the implementation method is: when the network is in normal condition, the service takes a short path.
  • the nodes on both sides of the fault point first perform a wrapping switch, and at the same time perform a re-topography, and the source node of the service obtains new network topology information. After that, the route is calculated according to the topology information, and a short path is selected for transmission.
  • This method can not only improve the protection speed of the network, but also improve the network bandwidth utilization.
  • the Steering protection method requires network topology, information, and network failures, and needs to wait for a re-topology discovery before finding the best route.
  • the virtual MPLS ring network of the present invention may have a network topology automatic discovery function, but is not required.
  • topology auto-discovery is to automatically discover when the network topology changes, and send new topology information to each node on the ring in time.
  • the topology discovery function of the MPLS ring network can be implemented by defining an MPLS 0AM packet.
  • the topology information can be spread to neighboring nodes in the form of 0AM packets.
  • the topology information is spread to the adjacent 5" points in the Tunnel LSP. 01580
  • a dedicated LSP can be defined for transmitting topology information between adjacent nodes.
  • the present invention utilizes the EXP (test) field in the MPLS label (this field is not defined in the standard, including 3bi t, and f is used as a priority, and 8 types can be identified.
  • EXP test
  • this field is not defined in the standard, including 3bi t, and f is used as a priority, and 8 types can be identified.
  • Priority Different LSPs can define different QoS parameters to make the supported SLAs (service level agreements) richer. By scheduling the LSP granules according to the QoS parameters, better guarantee and support for differentiated QoS. Specific implementation steps: ⁇ ⁇ :
  • the value is determined according to a certain algorithm, for example, according to the pri (priority) domain of the VLAN (Virtual LAN: LAN) service (the domain contains 3bi t, which can represent 8 priority 1 And / or IP (Internet Protocol) service T0S (service type) domain (the domain identifies the type of IP service, such as video services, etc.) and / or the priority specified by the administrator and other information to classify the traffic. You can choose one or more kinds of information to be combined according to your needs. It is a video service for female ⁇ for T0S. When VLAN is also high priority, it is defined as first priority; when VLAN is lower priority, It is defined as the second priority and so on.
  • the ring network uses an intrinsic mechanism (depending on the fair algorithm used) to implement a fairness algorithm to control the utilization of bandwidth.
  • the fairness algorithm is a mechanism that allows each user on the ring to enjoy bandwidth fairly. It does not assign a fixed limited bandwidth to each user as SDH, but the entire bandwidth on the 4th ring is allocated to the user as a global resource. Each node can learn the amount of data allowed to be sent to the ring based on the results of the fairness algorithm.
  • the ring network of the present invention not only has all the advantages of the RPR ring network, but also has more advantages than the RPR ring network, for example: simple service processing, high efficiency, and realization of cross-ring end-to-end service provision, multi-ring Service interworking at the time of intersection/tangentiality, and the supported SLA (Service Level Agreement) is more abundant, and the 0AM function in MPLS can be fully utilized.
  • simple service processing, high efficiency, and realization of cross-ring end-to-end service provision multi-ring Service interworking at the time of intersection/tangentiality
  • SLA Service Level Agreement

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Small-Scale Networks (AREA)

Description

一种环网及其业务实现方法 技术领域 本发明涉及网络通信技术领域, 具体涉及一种环网及其业务实现方 法。
背景技术
随着 Internet (因特网) 的蓬勃发展, 人们对网络的应用也提出了更 高的需求。 MPLS (多协议标签交换)技术就是为了解决 IP网络中报文的 快速转发及 QoS (服务质量)保证和流量工程的问题而发展的一种新技术。 它吸收了 ATM (异步传输模式) 的 VPI/VCI (虚通道标识 /虚电路标识) 交换的一些思想, 无缝地集成了 IP路由技术的灵活性和二层交换的简捷 性, 在面向无连接的 IP网络中增加了 MPLS这种面向连接的属性。 通过 采用 MPLS建立 LSP (标签交换路径) 的方法为 IP网增加了一些管理和 运营的手段。 伴随着近年来数据业务需求的快速增长, MPLS技术以其在 多业务 /协议支持能力、 MPLS L3/L2 VPN (虚拟专用网)和流量工程等各 方面的优异表现, 逐步成为数据网络解决方案的主流技术, 并进一步从网 络的核心层向汇聚层和接入层拓展。
MPLS协议的关键是引入了标签(Label )的概念。 它是一种短的易于 处理的、 不包含拓朴信息、 只具有局部意义的信息内容。 Label短是为了 易于处理, 通常可以用索引直接引用。 只具有局部意义是为了便于分配。 MPLS技术的基本特征是: 对于每一个进入 MPLS域的报文, 根据报文的 流分类, 边缘路由器给它分配一个标记, 自此, 所有的 MPLS路由器收到 这种打上标记的报文后直接按标记进行转发, 同时将标记更改成自己与下 一个路由器约定的新标记, 如此一直到带标记的报文离开 MPLS域后, 重 新以 IP报文出去。 这种带标记的报文离开 MPLS域时, 标签被边缘路由 器分离。 MPLS数据包的服务庸量类型可由 MPLS边缘路由器根据 IP包的 各种参数来决定, 如 IP的源地址、 目的地址、 端口号、 服务类型 (TOS ) 值等参数。 如对于到达同一目的地的 IP包, 可根据其 TOS值的要求来建 立不同的转发路径, 以达到其对传输质量的要求。 同时, 通过对特殊路由 的管理, 还能有效的解决网络中的负载均衡和拥塞问题。 如当网络中出现 拥塞时, MPLS可实时地建立新的转发路由来分担其流量, 以緩解网络拥 塞。
目前业界主流的 MPLS封装是 Martini封装, 在 I^TF草案 draft- martini- 12circuit- encap-mpls-Q4中定义。 Martini封装的 MPLS帧, 包含两层标签, 外层为 Tunnel label (隧道标签), 内层为 VC label (虛容器标签)。 标记栈 可以无限嵌套, 从而提供无限的业务支持能力。
如图 1所示: 其中, 隧道 LSP1为 R1和 R5之间的一个管道 LSP, 标签交 换路径为 10 ( R1/R6 )→20 ( R6/R5 ) , R6只完成隧道标签交换; 虚容器标 签 LSP2为 R1和 R5之间的一条业务 LSP, 标签交换路径为 10.01 ( R1/R6 ) →20.01 ( R6/R5 ) , R6只完成隧道标签交换, 内层标签对 R6不可见。
MPLS LSP通常采用 1+1或 1:1保护技术, LSP的保护是一种端到端 的保护。 这种保护方式虽然能够有效地完成业务的保护, 但还存在一定的 缺点:
A )对于 MPLS网络, 如果对网络中所有节点之间的业务都进行 1 : 1 或者 1+1保护, 则需要在所有节点的任意两点之间分别建立一条工作 LSP 和一条保护 LSP, 即存在工作 LSP的 mesh连接和保护 LSP的 mesh连接, 工作 LSP和保护 LSP都存在 N平方的问题, 导致在网络节点比较多时, 网络需要配置和维护大量的 LSP信息;
B )使用 MPLS OAM功能实现保护, 属于被动式保护, 即由业务的 宿端持续检?则 CV、 FDI等 OAM帧, 连续发现 CV丢失或者多次收到 FDI 错误, 则认为 LSP失效, 考虑到 OAM为带内传送, 会占用 LSP带宽, 所 以 OAM帧不可能实时发送, 发送频率也不易过快, 导致故障发现不够及 时, 无法达到电信级保护倒换需求;
C ) 1+1保护为双发选收方式, 导致网络带宽利用率较低。
为了提供更好的网络体系结构, 使业务得到更好的 QoS (服务质量) 和保护功能。 RPR (弹性分组数据环)技术提供了一个很好的解决方案。
RPR采用双环结构: 每对节点之间都有两条路径, 保证了高可用性; 对环路带宽采用空间重用机制, 单播数据传送可在环的不同部分同时进 行, 提高了环路带宽的利用率。 RPR环网可以支持 50ms的快速保护: RPR环网可采用两种保护机制 , 一种是源路由方式(Steering ), 即直接在业务的源点进行倒换, 可保证业 务走最佳路径; 一种是在发生故障的两个节点进行 ^回 (Wrapping ) 的方 式,类似于 SDH(同步氣字体系)的 2纤 MS-SPRing(复用段共享保护环)。
RPR技术是一种在环形结构上优化数据业务传送的 MAC (媒质接入 控制层)层协议, 能够适应多种物理层, 可有效地传送话音、 图像等多种 业务类型的数据。 它融合了以太网技术的经济性、 灵活性、 可扩展性等特 点, 同时吸收了 SDH环网的 50ms快速保护的优点, 并具有网络拓朴自动 发现、 环路带宽共享、 公平分配、 严格的 COS (业务分类)等功能。
但 RPR技术也有其 用的局限性, 由于 IEEE802.17规范的是专为单 个物理环或逻辑环 (跨多个 SDH物理环的虚容器 VC通道构成) 而设计 的 RPR MAC层技术, 因此 RPR的应用仅局限在单环, 跨环时必须终结, 即无法实现跨环业务的端到端带宽共享、 公平机制、 QoS (服务质量)和 保护功能。
目前, 为了解决 RPR环在相交、 相切时环之间的业务互通问题, 一
^:的方法是在入、 出环处, 多环之间, 引入二层或三层交换机, 这使得网 络比较复杂, 网络结构也不清晰。 另一种解决方案是采用 MPLS over RPR (弹性分组数据环承载多协议标签交换)来弥补 RPR的缺点, 但这样处 理, 就引入了两层: RPR层和 MPLS层, 如图 2所示, 这样就会增加业务 处理的复杂性, P争低处理效率; 同时, 由于 RPR环网的帧格式是专用格 式, MPLS层的引入会增加每个数据包的开销, 使得带宽的利用率降低; 而且只能基于物理链路或者子通道构建 RPR保护环, 导致组网不够灵活, 提供业务较慢, 不宜构建大型网络。
发明内容 本发明的目的是提供一种环网及其业务实现方法, 以克服现有技术中
RPR环网的缺点, 解决跨环业务的互通, 提供更好的 QoS保护和丰富的 SLA (服务等级协议)。
本发明的目的是通过以下技术方案实现的:
一种环网, 包括: 处于同一逻辑层上的多个节点, 用于将业务接入所 述环网或从所述环网上接收业务; 各相 4(5节点之间通过两条方向相反的虚 拟通道相连, 所述虚拟通道用于承载适 ί业务数据; .物理链路, 用于承载 适配到所述虚拟通道的业务数据。
优选地, 所述虚拟通道为基于多协、议标签交换的标签交换路径。 所述虚拟通道还可为异步传输模 虚拟通道连接。
所述节点将承载的适配到所述虚 4 通道的业务数据通过以太网媒体 接入控制协议或通用成帧规程或链路接入规程或高速数字链路控制连接 适配到所述物理链路中。
所述物理链路包括: 以太网链路、 同步数字网链路、 光传送网链路、 虚级联组。
所述物理链路位于环网和 /或相交坏和 /或 mesh网中。
所述业务包括: 因特网业务、 以;^网业务、 异步传输模式业务。 优选地, 所述节点接入或接收的业务数据的封装格式是多协议标签交 换的标准帧格式。
一种环网的业务实现方法, 所述环网包括: 处于同一逻辑层上的多个 节点, 两条方向相反连接各相邻节点的虚拟通道和用于承载所述适配到所 述虚拟通道的业务数据的物理链路; 述方法包括:
A、 在各节点配置所述虚拟通道构成东、 西双向环网;
B、 在各节点建立业务标签交换路径调度表;
C、 在业务源节点, 根据所述标签交换路径调度表获取业务宿节点;
D、 根据预定的调度算法将所述业务调度上所述环网;
E、 将不同节点的业务以标签交换 各径的方式复用在同一个虚拟通道 中进行传送;
F、 在所述业务宿节点将所述业务"周度下所述环网。
所述方法还包括: 对所述环网采用环倒换保护和 /或源路由保护。 所述步骤 A还包括: 绑定所述虛 通道与对应的环网。
所述步骤 B 包括: 静态配置并维:护和 /或动态配置并维护所述业务标 签交换路径调度表。
所述动态配置并维护所述业务标签交换路径调度表包括: 利用标签分 发协议及资源预留协议配置并维护所述标签交换路径调度表。
所述标签交换路径调度表包括: 标签动作信息 目的端口信息。 所述步驟 C还包括:按预定规则完成非多协议标签交换业务的多协议 标签交换封装。
所述预定规则包括:
按目的地址将非多协议标签交换业务分组划分沩不同的转发等价类 , 根据分組所属的转发等价类, 将适当的标签插入分纽头中, 完成多协议标 签交换的封装;
按服务质量要求将非多协议标签交换业务分组划分为不同的转发等 价类, 根据分组所属的转发等价类, 将适当的标签插入分组头中, 完成多 协议标签交换的封装。
所述预定的调度算法包括: 严格优先级调度方法。
所述方法还包括: 对环内业务和跨环业务采用业务标签交换路径的
1 :1和 /或 1+1保护。
所述方法还包括:
在源节点 , 对本地上环业务进行流分类;
根据所述流分类等级填充多协议标签交换中 ^试验域; 以及, 根据所述试—险域指示的不同优先級, 将所述本她上环业务调度到不同 的出端口队列上环网传输。
所述流分类的信息包括:虚拟局域网业务的优先级域和 /或因特网协议 业务的业务类型域和 /或管理者指定的优先级。
所述方法还包括:
在所述环网上的相邻两节点之间建立专用标签交换路径传送公平算 法十办议信息;
由所述环网上的每个节点的媒质接入控制层现测与本节点相连的链 路的利用情况, 并将观测到的情况通知环网上的^ t有节点; 以及,
所述环网上的每个节点根据所述公平算法协议及获得的通知调节本 节点向所述环网发送数据的速率。
由以上本发明提供的技术方案可以看出, 本发明基于 MPLS技术, 使 环网不仅具备了 RPR环网的所有功能, 而且比 RPR环网具有如下优 : 不依赖于特定的物理层技术, 可以在多种物理层技术实现的网络中构建本 发明环网; 不依赖物理网络拓朴而构建虚拟 /物理或者逻辑的 MPLS 不网, 实现更加灵活、 快捷, 可以有效提高业务的开通速度, 尤其是 VPN ( 拟 专用网)业务; 该虚拟 MPLS网的层次比 RPR少了一个层次, 因此 处 理上更筒单, 使业务的配置更简单清晰, 提高业务的保护速度和效率; 该 网采用标准的 MPLS帧格式, 使业务对环不具有依赖性, 不需要其他 4t助 技术即可在本发明环网上实现跨环的端到端业务供给以及多环相交 / 目切 时的业务互通, 同时提高了网络带宽的利用率; 本发明环网上不同的业务 LSP都可以定义不同的 QoS参数, 支持的 SLA更丰富, 通过对 LSP颗粒 按事先协商的 QoS参数进行调度, 可以更好地保障和支持差异化的 QoS; 由于在该环网上业务的承载方式是 LSP, 属于面向连接的技术, 可以充分 利用 MPLS中的 OAM (操作与维护)功能: 包括 LSP连通性校验 ( CV )、 LSP快速失效检测 (FFD )、 前向缺陷指示(FDI )、 后向缺陷指示(BDI ) 等, 对于业务进行有效的检测和维护。
附图说明 图 1是 MPLS网络示意图;
图 2是现有技术中 RPR环网的层次结构示意图;
图 3是本发明环网的拓朴图;
图 4是本发明环网的层次结构示意图;
图 5是图 4所示环网以 MPLS LSP作为虚拟通道复用层时的结 示意 图;
图 6是本发明环网的不同物理组网方式示意图;
图 7是图 6所示网络通过配置 LSP组成虚拟 MPLS环网的简化结构; 图 8是以 SDH/SONET作为本发明环网的承载层时的网络拓扑图; 图 9是图 8所示环网的层结构示意图;
图 10是以以太网作为本发明环网的承载层时的网络拓朴图; 图 11是图 10所示环网的层结构示意图;
图 12是本发明环网的业务实现方法的流程图; 图 13是本发明中环网业务上下的实现示意图;
图 14是本发明中跨环业务上下的实现示意图; f
图 15是本发明中跨环业务的保护实现示意图;
图 16是本发明中采用环倒换方式实现基于网络的保护的过程; 图 17是本发明中采用源路由方式实现基于网络的保护的过程。
具体实施方式 本发明的核心在于利用 MPLS技术构建不依赖于物理拓朴结构和具 体物理层技术的环网技术, 虚拟 MPLS环网, 由处于同一某逻辑层次上的 多个节点组成双通道双向环型网絡。 该虚拟环网拓朴结构如图 3所示, 包 括:
处于同一逻辑层上的多个节点, 用于将业务接入环网或从环网上接收 业务; 各相邻节点之间通过两条方向相反的虚拟通道相连, 用来传输节点 间的业务数据; 业务数据可以采用不同的物理链路承载。
两条虚拟通道组成两个方向相反的环: 西向环和东向环, 西向环顺时 针传送业务, 东向环逆时针传送业务, 环上节点接入或接收的业务采用标 准的 MPLS帧结构, 不需要重新定义帧格式, 虚拟通道以 LSP (标签交换 路径)的方式承载不同节点的业务数据。 各节点通过业务的 MPLS标签识 别是否为本地业务, 如果是本地业务则将业务剥离 , 否则通过虚拟通道传 送到下一个相邻站点。
该虚拟通道可以通过多种技术构建, 比如, 通过标签交换路径构建, 还可以通过其他 VPN (虚拟专用网)专线技术构建, 如 ATM的 VPC (虚 通路连接)等。
该虚拟环网网络层结构如图 4所示:
其中, MPLS业务汇聚层采用 MPLS技术, 通过业务 LSP实现业务的 接入、传送、汇聚和复用。虚拟通道复用层可以釆用但不限于 MPLS技术, 以 LSP作为构建本发明虚拟 MPLS环网的虚拟通道, 或者说作为 MPLS 环网承载层时, 网络层结构如图 5所示。 虚拟通道层可以省略, 在这种情 况下,虚拟 MPLS环网的承载层变为数据链路层,从而虚拟 MPLS环网演 变成一种基于物理链路或者逻辑子通道的 MPLS环网, 如基于 GE (千兆 以太网)连接的 MPLS环网, 或者基于 VCG (虚级联组)连接的 MPLS 环网。
业务层可以接入各种不同的业务, 如 IP (因特网)业务、 ETH (以太 网)业务、 ATM (异步传输模式)业务等。
MPLS 业务层采用 MPLS 技术进行业务的汇聚和复用, 使用标准的
MPLS帧格式, 以保证和现存的 MPLS网络的业务互通, 也保证了跨环业 务的互通。
数据链路层可以但不限于以太网 MAC (媒体接入控制)、 GFP (通用 成帧协议)、 LAPS (链路接入规程)、 HDLC (高级数据链路控制)等数据 链路层技术, 即环网上的节点可以将虚拟通道上传送的业务数据通过以太 网 MAC协议、 GFP、 LAPS、 HDLC等连接适配到物理链路中。 本发明环 网对数据链路层没有依赖性。
本发明中环网对物理层也没有依赖关系, 可以根据需要采用不同的物 理层技术。 如可以采用以太网技术、 POS ( Packet over SDH , 基于传输网 的包交换)技术或 EOS ( Ethernet over SDH, 基于传输网的以太网)技术 的 VCAT (虚级联)技术。
POS是 IP数据包通过采用 PPP (点到点协议 )对 IP数据包进行封装, 并采用 HDLC的帧格式映射到 SDH/SONET (同步数字体系 /同步光网络) 帧上, 按某个相应的线速进行连续传输, 它保留了 IP面向非连接的特性。 其中, PPP 协议提供多协议封装、 差错控制和链路初始化控制等功能; HDLC帧格式负责同步传输链路上 PPP封装的 IP数据帧的定界。 PPP协 议可将 IP数据包切成 PPP帧以满足映射到 SDH/SONET帧结构上去的要 求。
EOS主要定义了将以太网帧进行封装后再映射到 SDH/SONET的 VC (虚容器)中的映射方法,位于以太网 MAC层与物理层的 SDH间作为数 据链路适配层。 主要封装映射方式有 PPP/HDLC、 LAPS和 GFP几种。
本技术领域人员知道, 传统的 MPLS LSP是一种端到端的 "连接"或 者说 "虚通道", 而以 MPLS LSP作为连接环网上各节点的虚拟通道, 并 为这些通道分配一定的带宽, 即形成了虚拟 MPLS环网结构, 也就是本发 明的环网。 这种方式和物理组网形式无关, 可以跨越物理网络的限制。 物 理组网可以是环网、 相交环、 mesh (网状) 网等。
如图 6所示的本发明环网的不同物理组网方式:
可以将 R6/R8之间的 Tunnel LSP1(经过 R2、R1 ) R8/R7之间的 Tunnel LSP3 (经过 R5、 R4 )和 R6/R7之间的 Tunnel LSP5 (经过 R3 )通过管理 软件(手工或者自动) 配置成西向环 Loopl。 同时, R6/R8之间的 Tunnel LSP2 (经过 R2、 Rl )、 R8/R7之间的 Tunnel LSP4 (经过 R5、 R4 )和 R6/R7 之间的 Tunnel LSP6 (经过 R3 )通过管理软件(手工或者自动)配置成东 向环 Loop2。 以 Tunnel LSP构建虚拟 MPLS环网的一个好处就是, 不受端 口资源的限制, 可以在网络中, 根据业务情况构建足够多的 MPLS环网。 如在现有的 MPLS网络中, 在带宽资源允许的情况下, 只需要在相关节点 之间配置几个 LSP作为虚拟通道,就可以快速开通这几个节点之间的具有 自愈保护功能的业务。 业务提供筒单、 快速。
该图中的虚拟 MPLS环网可以简化成图 7所示结构, R6、 R7和 R8 三个网元虽然在物理上不是相邻的,但是可以通过通道 LSP1〜LSP6,将它 们配制成一个虚拟的 MPLS Ring, LSP1、 LSP3和 LSP5构成西向环, 顺 时针传送业务, LSP2、 LSP4和 LSP6构成东向环, 逆时针传送业务。 三 个网元之间的业务, 通过业务 LSP承载到通道 LSP中。
前面已经提到, 本发明环网的虚拟通道复用层不是必需的, 这时, MPLS 环的承载层变成了数据链路层, 通道可以是物理链路或者是子通 道, MPLS环的通道带宽不再被复用。
例如, 在 SDH/SONET网络中, 可以以 VCG作为通道, 构建一个双 通道双向网型网络,各相邻节点之间通过两条方向相反的 VCG通道相连, 组成方向相反的两个 VCG环, 分别称为西向环和东向环, 西向环顺时针 传送业务, 东向环逆时针传送业务, 环上传送的业务以 MPLS标签交换路 径的方式 7 载。
如图 8所示: VCG1、 VCG3、 VCG5和 VCG7构成西向环, 顺时针传 送业务, VCG2、 VCG4、 VCG6和 VCG8构成东向环, 逆时针传送业务。 这种技术的环网层结构如图 9所示。 其中, SDH和虚级联共同组成虚拟 MPLS环网网络的物理层, 数据链路层采用 GFP或者 LAPS/HDLC构成 MPLS环网的承载层。
也可将以太网直接作为虚拟 MPLS环网的承载层, 比如, 采用路由器 的 GE端口, 将各节点连接成一个双纤双向环形结构, 西向环顺时针传送 业务, 东向环逆时针传送业务, 如图 10所示。 这种技术的环网层结构如 图 11所示。 其中, 以太网的 MAC层组成虚拟 MPLS环网网络的物理层, 数据链路层采用 GFP或者 LAPS HDLC构成 MPLS环网的承载层。
为了使本技术领域的人员更好地理解本发明方案, 下面结合图 12所示 的流程详细说明利用该环网传送用户业务的流程, 包括以下步骤:
S11 : 在各节点配置虚拟通道构成东、 西双向环网, 即在同一某逻辑 层上构建虚拟通道, 为其分配一定的带宽, 使其构成东向环和西向环。 虚 拟通道的构建可以采用多种技术, 比如, 采用 MPLS LSP在 MPLS业务汇 聚层和数据链路层之间构建虚拟通道, 还可以基于物理链路或者逻辑子通 道在物理层构建物理或者逻辑子通道,从而使虚拟的 MPLS环网演变成基 于物理链路或者逻辑子通道的 MPLS环网,如基于 GE连接的 MPLS环网, 或者基于 VCG连接的 MPLS环网。
步驟 S12: 绑定所述虚拟通道与对应的环网。 在实际物理网络中, 可 能会构建多个虚拟 MPLS环, 每个端口也可能包含多个虚拟通道, 因此, 需要将构建的每条虚拟通道与其对应的环网绑定。 这样就由物理网絡中的 多个相邻或不相邻的端口通过构建的虛拟通道组成了一个完整的环路。 业 务层接入的不同业务首先要通过 MPLS封装适配到该虚拟通道复用层, 然 后, 再通过数据链路层的封装适配到物理层。
步骤 S13: 在各节点建立业务标签交换路径调度表。 该标签交换路径 调度表包含标签动作和目的端口等信息, 标签动作包括弹出标签, 加入标 签, 交换标签, 表的建立方式可以采用静态配置的方式, 也可以是用 LDP (标签分发协议)协议、 RSVP (资源预留协议)协议创建和维护, 或者 采用两者相结合的方式。
步驟 S14: 在业务源节点入口, 按预定规则完成非多标签协议交换业 务的多标签协议交换封装。其详细过程可参照图 13 ,该环网包括四个节点 A、 B、 C、 D。 在源节点 A入口, 将非 MPLS的业务分组, 按照预定规则 将 MPLS业务分组划分为不同的转发等价类,根据分组所属的转发等价类, 将适当的标签插入分组头中, 即完成多协议标签交换的封装。 所述预定规 则由用户指定, 比如可以按照目的地址、 QOS (服务质量)要求等对非 MPLS业务分组。 如果是 MPLS业务, 则不需要此步骤。
步骤 S15: 在业务源节点根据业务标签交换路径调度表获取业务宿节 点。 具体为: 根据业务标签交换路径调度表, 进行标签动作 (弹出标签, 加入标签, 交换标签)后, 查找到出端口, 即获取业务宿节点。
步骤 S16: 根据预定的调度算法将业务调度上环网。 比如, 可以按照 严格优先级调度算法将用户业务调度上环网, 以便为用户提供更好的等级 服务。
步驟 S17: 将不同节点的业务以标签交换路径的方式复用在同一个虚 拟通道中进行传送。 虚拟通道中复用的业务可以通过不同的物理链路承 载, 比如, 前面提到的 SDH/SONET 的 VCAT , 将复用业务通过 GFP/LAPS/HDLC等封装映射到 SDH/SQNET的 VC (虚容器) 中; 或者 将以太网直接作为承载复用业务的物理链路,通过路由器的 GE端口传输。
步骤 S18:经过中间节点进行标签交换将用户业务传送到业务宿节点。 比如图 13中, 经过中间节点 B进行标签交换后迅速前传至下一节点<3。
步骤 S19: 在业务宿节点将业务调度下环网。参照图 13,在宿节点 C, 经查业务标签交换路径调度表, 获悉此节点为该标签交换路径的宿节点, 用户业务被调度下该环网, 进行客户层信号的进一步处理。
本发明环网采用标准的 MPLS帧格式传送用户业务,使业务对环不具 有依赖性, 因此, 在该环网以外的设备不需要进行任何处理即可识别该环 网的用户帧,从而方便地实现跨环的端到端业务供给、多环相交 /相切时的 业务互通。
图 14为跨环业务在本发明环网上、 下的实现示意图:
一个跨本发明环网 A、环网 B的业务,环网 A包括四个节点:节点 A、
B、 C、 D; 环网 B包括四个节点: 节点 E、 F、 G、 H。 411是该业务在环 网 A上的实际路径, 412是该业务在环间上的实际路径, 413是该业务在 环网 B上的实际路径。 而路径 411、 412、 413实现时分别对应 1条 LSP, 即在环网 A上, 从节点 D到节点 A, 是一条 LSP:LSP1 ; 在环网 B上, 从 节,^ E到节点 H, 是一条 LSP:LSP3; 在环间, 从节点 A到节点 E, 则是 另夕卜一条 LSP:LSP2,在节点 A和节点 E分别完成 LSP1到 LSP2标签交换、 LSP2到 LSP3的标签交换。
由此可以看出该跨环业务在环网上和环网间, 都可以利用 MPLS 的 LSP调度来实现业务调度,不需要做其他的转换即可完成跨环业务的调度。
多环相交、相切时,业务跨环传送的过程与上述类似,在此不再赘述。 为了更好地保障网络的正常运行, 本发明还对该环网采取了如下的多 种保护措施:
1、 基于业务 LSP (标签交换路径的保护) 的保护
( 1 )环内业务的保护:利用 MPLS OAM (操作与维护)功能实现 LSP 的 1 :1或 1+1保护。
具体实现为: 工作 LSP和备用 LSP逆向配置(如: 工作 LSP配置在 西向环中, 保护 LSP配置在东向环中); 在工作 LSP出错后, 利用 MPLS 的 OAM功能, 及时监测到该情况; 然后启用备用的 LSP。
( 2 )跨环业务的保护: 利用 MPLS OAM功能, 实现 LSP的 1:1或 1+1保护。
如图 15所示, 具体实现为: 工作 LSP和备用 LSP跨越相交环的不同 链路;在工作 LSP出错后,利用 MPLS的 OAM功能,及时监测到该情况, 启用备用的 LSP。
在上面提到 MPLS的 0AM功能, 下面简单介绍如下:
MPLS的 0AM (操作和维护)主要由 ITU-T Rec. Y. 1711定义的 MPLS OAM 帧来保证, 目前定义的帧类型有 CV (连通性校验)、 FDI (前向缺陷指示)、 BDI (后向缺陷指示)、 性能报文、 环回请求、 环回响应, 但是定义了具体 格式和操作规程的只有 CV、 FDI , BDI三种。
( a )连通性校验: CV流在 LSP的源 LSR产生, 以 1/S的速度发送,在 LSP 的^ LSR终结; CV报文携带网络唯一标识符(TTSI ) , 这样奠定了能检测 所有缺陷的基础。 ( b )前向缺陷指示: FDI报文的产生是作为对检测故障的行为的响应 (例如来自于 CV流 ή 缺陷), 它的主要目的是压制检测到错误的层以上的 层网络的告警;
( c )后向缺 |¾指示: BDI流在返回通路(比如一个返回的 LSP )上被 插入, 用来将在下行 LSP的 LSR (标签交换路由器)宿点中检测到的缺陷通 知给上行 LSR (前向 LSP的源点) 。
2、 基于网络的保护
传统环网的保护一般具有 wrapping (环倒换保护)和 s teer ing (源 路由保护) 两种。
本发明虚拟 MPLS 环可以才艮据节点上下业务的方式, 分为单向保护环 和双向保护环。 前者指定两个方向相反的环中的一个用作工作通道, 用于 在正常情况下传送 务, 业务的上下始终在工作通道上完成, 另外一个环 作为保护通道使用, 用于在网络故障中保护工作通道; 后者则可以同时在 两个方向上传送业务, 正常情况下, 业务走短路径, 故障发生后, 业务通 过长路径传送。
MPLS环的保护可以采用 Wrapping和 s teering的保护方式,下面对这 两种保护方式分别 "^细说明。
( 1 ) 环倒换保护 (wrapping ):
Wraping保护可以不需要信令的帮助, 只需要检测服务层的通道链路 状态, 当服务层出现 SF (信号失败)时, 对应于故障点一侧的通道进行工 作通道和保护通道之间的桥接。
具体实现过程 下:
I. 环网节点间通过信令交换拓朴信息, 每个节点都知道网络的 状态 (可选);
I I. 当环网光纤中断后, 光纤中断处两端节点会监测到缺陷信息 和位置;
I II. 光纤中断处两端节点会发出控制信令沿光纤方向通知各个节 点 (可选);
IV. 在缺陷发生的相邻节点,分别将业务环回到另外一个虚拟通道 环进行传送;
V. 在环保护切换时, 按业务流的不同服务等级的高低, 依次向反 向通道倒换业务。
图 16示出了以 LSP为承戴层的单向保护环倒换的实现过程: 其中, LSP1、 LSP3、 LSP5、 LSP7、 LSP9和 LSP11构成 MPLS Ring的西 向环, 作为工作通道, 顺时什传送业务。 LSP2、 LSP4、 LSP6、 LSP8、 LSP10 和LSP12构成MPLS Ring的东向环,作为保护通道,逆时针传送业务。 LSPQ 是从 R1到 R4节点的业务, JE常情况下顺时针经过 R2 、 R3, 传送到 R4 站点时, 被目的站点 R4剥离。 当 LSP3或 /和 LSP4通道发生故障时, R2 和 R3 都在邻近故障点一侧 ^寄工作通道和保护通道进行桥接。 这时对于 LSPQ本应在 R2节点被封装 ll通道 LSP3中发往 R3, 故障发生后, 则被封 装到保护通道 LSP2中逆时针发送到 R1 , 经 R6、 R5、 R4的保护通道到达 R3, 在 R3将其倒换到工作通道并封装在通道 LSP5中, 顺时针发送到 R4 , R4发现是本地业务后将其剝离。保护通道在网絡正常下,可以传送额外业 务。
( 2 ) 源路由保护 ( steer ing ):
s teer ing保护则需要信令的协助,一般在正常情况下,业务走短路径, 网络故障后, 通过拓朴发现, 重新计算业务路由。
具体实现过程如下:
I. 环网节点间通过信令交换拓朴信息,每个节点都知道网络的状 态;
I I. 当环网光纤中断后,光纤中断处两端节点会发出控制信令将受 影响的业务 LSP的信息通知其它节点;
I I I. 相应的业务 LSP的源节点接收到信息后,立即将此业务 LSP改 向另外一个虚拟通道环进行传送, 从而实现源路由保护。
图 17示出了以 LSP为 7 载层的双向保护环的源路由保护实现过程: LSP1、 LSP3、 LSP5、 LSP 7, LSP9和 LSP11构成 MPLS Ring的西向环, 顺时针传送业务。 LSP2、 LSP4、 LSP6、 LSP8、 LSP10和 LSP12构成 MPLS Ring 的东向环, 逆时针传送业务。 LSPQ和 LSPN为 R1和 R3节点之间的两条业 0
- 15- 务, 正常情况下 LSPQ顺时针经过 R2 到 R3, 并被 R3剥离。 LSPQ逆时针经 过 R2 到 Rl, 并被 R1剥离。 当 LSP3或 /和 LSP4通道发生故障时, 通过拓 朴重新发现, R1发现 LSPQ不可达, 将 LSPQ倒换东向环上, 逆时针将业 务通过 R6、 R5、 R4 , 传送到 R3, 并被 R3剥离。 R3发现 LSPN顺时针不可 达, 则将 LSPN倒换西向环上, 顺时针将业务通过 R4、 R5、 R6 , 传送到 Rl, 并被 R1剥离。 在 Steer ing保护方式下, 两个环上都应该预留一定的保护 带宽以备故障时保护工作业务。 预留带宽在网洛正常情况下, 可以传送额 外业务。
Wrapping保护方式倒换速度快,可以不需要拓朴信息,但是由于倒换 发生时, 业务在网络中进行缠绕, 导致带宽利用率较低。 而 Steer ing保 护方式, 带宽利用率高, 但是需要网络拓朴信息, 网络故障时, 需要等待 重新的拓朴发现, 然后才能够找到最佳路由, 所以倒换时间较长。
为了避免 wrapping方式的带宽利用低和 St eer ing的倒换速度较长的 问题, Wrapping + Steering方式提供了较好的解决方案。 实现方法是, 网络正常情况下, 业务走短路径, 网络故障发生时, 故障点两侧的节点首 先进行 wrapping倒换, 同时进行重新拓朴发见, 待业务的源节点得到新 的网络拓朴信息后, 根据拓朴信息计算路由, 选择一个短路径进行传送。 这种方式既可以提高网络的保护速度, 又可以提高网络带宽利用率。
前面提到 Steering保护方式需要网络拓 ^、信息, 网络故障时, 需要 等待重新的拓朴发现, 然后才能够找到最佳路由。 本发明虚拟 MPLS环网 可以具备网络拓朴自动发现功能, 但不是必需的。
拓朴自动发现的功能是当网络拓朴发生变化的时候能够自动发现, 并 及时将新的拓朴信息发送到环上各节点。 MPLS环网的拓朴发现功能可以通 过定义一种 MPLS 0AM报文来实现。
a)在以 MPLS Tunnel LSP作为承载层的 MPLS 环网中, 拓朴信息可以 使用 0AM报文的形式向相邻节点扩散, 目前 MPLS 0AM帧制定了 5 种类型, 可以定义一种新的 0AM帧, 用来7 载环网的拓朴信息, 在 Tunnel LSP中将拓朴信息扩散给相邻 5 "点。 01580
- 16- b)在以物理通道或者子通道作为^载层的 MPLS 环网 , 可以定义专 门的 LSP, 用于在相邻节点之间传送拓朴信息。
为了给用户业务提供更好的服务质量,本发明利用 MPLS标签中的 EXP (试验)域(该域在标准中是没有定义用途的,含 3bi t ,通 f用作优先级, 可标识 8种优先级), 不同的 LSP都可以定义不同的 QoS參数, 使支持的 SLA (服务级别协议) 更丰富, 通过对 LSP颗粒按事先协 的 QoS参数进 行调度, 更好地保障和支持差异化的 QoS。 具体实现步骤:^下:
I. 在源节点, 对本地上环业务进行流分类。
I I. 根据流分类等级填充 MPLS的 EXP域:
a ) 对于非 MPLS流进行 MPLS封装。 对于 MPLS格弍中的 EXP域, 根据一定的算法来决定其取值, 比如, 根据 VLAN (虛扛:局域网)业务 的 pri (优先級)域(该域含 3bi t , 可表示 8种优先 1 )和 /或 IP (因 特网协议)业务的 T0S (业务类型)域(该域标识 IP业务的类型, 如 视频业务等)和 /或管理者指定的优先级等信息对业务进行流分类。 可根据需要从中选择一种、 或多种信息进行组合, 比女 π对于 T0S是视 频业务的, 当 VLAN也是高优先级的, 就定义为第一 先级; 当 VLAN 是较低优先级的, 就定义为第二优先级等等。
b )对于 MPLS流, 根据需要, 选择沿用已有的 EXP域或重新指 定 EXP域。
c ) EXP 域填充, 只针对本地上环的业务, 且对环网所有的节点 采用相同的算法。
I II. » EXP域指示的优先级的不同,将用户业务调度到不同的出端 口队列。
IV. 采用某种算法和策略(比如严格优先级调度等调度算法), 调度 不同的优先队列到出端口, 即上环网上传输。 由于环网环上的带宽是共享资源, 极其容易被网络上的个别节点或个 別用户过度使用造成网络瘫痪。 因此, 在本发明环网中采用信令收集各节 点的业务带宽信息, 利用一定的公平算法, 控制各个站点的上环业务, 使 每个站点都能公平的访问环带宽。 具体步骤如下: -
I. 在环上的相邻两点之间建立一条专门的 LSP, 用于传送公平算法 协议信息;
I I. 每一个节点的 MAC层一直观测紧靠它的链路的利用情况, 然后把 这个信息告知环上所有的节点;
I I I.环网采用内在的机制(取决于所釆用的公平算法)执行公平算法 以控制带宽的利用。公平算法是一种可以让环上每一个用户公平享用带宽 的机制, 它不像 SDH给每一个用户分配一个固定的有限带宽, 而是 4巴环上 的整个带宽作为一个全局资源分配给用户。每一个节点可以才艮据公平算法 的结果获悉允许向环上发送的数据量。
IV. 建立一个反馈机制,根据上一步骤的结果调节源节点向网络中发 送数据的速率。 从而实现环带宽的公平访问。
由此可见, 本发明环网不仅具备了 RPR环网所有的优势, 而且比 RPR 环网具有更多的优势, 比如: 业务处理简单, 效率高, 实现跨环的端到端 业务供给, 多环相交 /相切时的业务互通, 支持的 SLA (服务等级协议)更 丰富, 可充分利用 MPLS中的 0AM功能等。
虽然通过实施例描绘了本发明, 本领域普通技术人员知道, 本发明有 许多变形和变化而不脱离本发明的精神, 希望所附的权利要求包括这些变 形和变化而不脱离本发明的精神。

Claims

权 利 要 求
1、 一种环网, 其特征在于, 包括: 处于同一逻辑层上的多个节点, 用于将业务接入所述环网或从所述环网上接收业务; 各相邻节点之间通过 两条方向相反的虚拟通道相连, 所述虚拟通道用于承载适配业务数据; 物 理链路, 用于承载适配到所述虚拟通道的业务数据。
2、 如权利要求 1 所述的环网, 其特征在于, 所述虚拟通道为基于多 协议标签交换的标签交换路径。
3、 如权利要求 1 所述的环网, 其特征在于, 所述虚拟通道为异步传 输模式虚拟通道连接。
4、 如权利要求 1或 2或 3所述的环网, 其特征在于, 所述节点将承 载的适配到所述虚拟通道的业务数据通过以太网媒体接入控制协议或通 用成帧规程或链路接入规程或高速数字链路控制连接适配到所述物理链 路中。
5、 如权利要求 1 所述的环网, 其特征在于, 所述物理链路包括: 以 太网链路、 同步数字网链路、 光传送网链路、 虛级联组。
6、 如权利要求 1或 5所述的环网, 其特征在于 , 所述物理链路位于 环网和 /或相交环和 /或 mesh网中。
7、 如权利要求 1所述的环网, 其特征在于, 所述业务包括: 因特网 业务、 以太网业务、 异步传输模式业务。
8、 如权利要求 1 所述的方法, 其特征在于, 所述节点接入或接收的 业务数据的封装格式是多协议标签交换的标准帧格式。
9、 基于权利要求 1 所述环网的业务实现方法, 所述环网包括: 处于 同一逻辑层上的多个节点, 两条方向相反连接各相邻节点的虚拟通道和用 于承载适配到所述虚拟通道的业务数据的物理链路; 其特征在于, 所述方 法包括:
A、 在各节点配置所述虚拟通道构成东、 西双向环网;
B、 在各节点建立业务标签交换路径调度表;
C、 在业务源节点, 根据所述标签交换路径调度表获取业务宿节点;
D、 ^^据预定的调度算法将所述业务调度上所述环网; E、 将不同节点的业务以标签交换路径的方式复用在同一个虚拟通道 中进行传送; "
F、 在所述业务宿节点将所述业务调度下所述环网。
10、 如权利要求 9所述的方法, 其特征在于, 所述方法'还包括: 对所 述环网采用环倒换保护和 /或源路由保护。
11、 如权利要求 9或 10所述的方法, 其特征在于, 所述步骤 A还包 括: 绑定所述虛拟通道与对应的环网。
12、 如权利要求 11所述的方法, 其特征在于, 所述步驟 B包括: 静 态配置并维护和 /或动态配置并维护所述业务标签交换路径调度表。
13、 如权利要求 12所述的方法, 其特征在于, 所述动态配置并维护 所述业务标签交换路径调度表包括: 利用标签分发协议及资源预留协议配 置并维护所述标签交换路径调度表。
14、 如权利要求 11 所述的方法, 其特征在于, 所述标签交换路径调 度表包括: 标签动作信息和目的端口信息。
15、 如权利要求 11所述的方法, 其特征在于, 所述步骤 C还包括: 按预定规则完成非多协议标签交换业务的多协议标签交换封装。
16、 如权利要求 15所述的方法, 其特征在于, 所述预定规则包括: 按目的地址将非多协议标签交换业务分组划分为不同的转发等价类, 根据分组所属的转发等价类, 将适当的标签插入分组头中, 完成多协议标 签交换的封装;
按服务盾量要求将非多协议标签交换业务分组划分为不同的转发等 价类, 根据分组所属的转发等价类, 将适当的标签插入分组头中, 完成多 协议标签交换的封装。
17、 如权利要求 11 所述的方法, 其特征在于, 所述预定的调度算法 包括: 严格优先级调度方法。
18、 如权利要求 11 所述的方法, 其特征在于, 所述方法还包括: 对 环内业务和跨环业务采用业务标签交换路径的 1 :1和 /或 1+1保护。
19、 如权利要求 11所述的方法, 其特征在于, 所述方法还包括: 在源节点, 对本地上环业务进行流分类; 根据所述流分类等级填充多协议标签交换中的试 域; 以及, 根据所述试验域指示的不同优先级, 将所述本地上环业务调度到不同 的出端口队列上环网传输。
20、 如权利要求 19所述的方法, 其特征在于, 所述流分类的信息包 括: 虚拟局域网业务的优先级域和 /或因特网协议业务的业务类型域和 /或 管理者指定的优先级。
21、 如权利要求 11所述的方法, 其特征在于, 所述方法还包括: 在所述环网上的相邻两节点之间建立专用标签交换路径传送公平算 法协议信息;
由所述环网上的每个节点的媒质接入控制层观测与本节点相连的链 路的利用情况, 并将观测到的情况通知环网上的所有节点; 以及,
所述环网上的每个节点根据所述公平算法协议及获得的通知调节本 节点向所述环网发送数据的速率。
PCT/CN2005/001580 2004-09-27 2005-09-27 Reseau en anneau et methode de mise en oeuvre de ses services WO2006034644A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US11/579,029 US20070280251A1 (en) 2004-09-27 2005-09-27 Ring Network And A Method For Implementing The Service Thereof
EP05792149A EP1796328A4 (en) 2004-09-27 2005-09-27 RING NETWORK AND METHOD FOR IMPLEMENTING ITS SERVICE

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200410080928.1 2004-09-27
CNB2004100809281A CN100334857C (zh) 2004-09-27 2004-09-27 一种环网及其业务实现方法

Publications (1)

Publication Number Publication Date
WO2006034644A1 true WO2006034644A1 (fr) 2006-04-06

Family

ID=36118584

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2005/001580 WO2006034644A1 (fr) 2004-09-27 2005-09-27 Reseau en anneau et methode de mise en oeuvre de ses services

Country Status (4)

Country Link
US (1) US20070280251A1 (zh)
EP (1) EP1796328A4 (zh)
CN (1) CN100334857C (zh)
WO (1) WO2006034644A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113785280A (zh) * 2019-03-27 2021-12-10 图核有限公司 环形计算机网络上的嵌入环

Families Citing this family (54)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1816035B (zh) * 2005-02-02 2010-07-07 华为技术有限公司 基于数据通信网的主备传输路径实现方法
US7623446B1 (en) * 2005-11-14 2009-11-24 Nortel Networks Limited MPLS virtual rings
CN101047632B (zh) * 2006-05-10 2010-09-22 华为技术有限公司 剥离rpr环网上环标识不匹配报文的方法及系统
US20100238813A1 (en) * 2006-06-29 2010-09-23 Nortel Networks Limited Q-in-Q Ethernet rings
CN100542122C (zh) * 2006-09-29 2009-09-16 华为技术有限公司 一种vlan交换隧道的复用方法和vlan交换域
CN101272310B (zh) * 2007-03-22 2013-02-13 华为技术有限公司 以太网环网自动保护倒换方法和装置
JP5076581B2 (ja) * 2007-03-23 2012-11-21 日本電気株式会社 光通信システムおよびノード装置およびサービスクラス設定方法
US7752338B2 (en) * 2007-03-30 2010-07-06 Extreme Networks, Inc. Ring topology discovery
JP2008306625A (ja) * 2007-06-11 2008-12-18 Nec Corp Vcat伝送システムおよびvcat帯域制御方法
US8913481B2 (en) * 2007-06-30 2014-12-16 Alcatel Lucent Method and system for efficient provisioning of multiple services for multiple failure restoration in multi-layer mesh networks
EP2223538B1 (en) * 2007-12-17 2016-03-23 Telefonaktiebolaget LM Ericsson (publ) Method and arrangement for network qos
CN101471693B (zh) * 2007-12-27 2013-04-17 华为技术有限公司 一种实现共享网格保护的方法和装置
CN101471849B (zh) * 2007-12-29 2011-04-06 华为技术有限公司 一种分组传送网的保护方法
JP5212469B2 (ja) * 2008-05-15 2013-06-19 富士通株式会社 コンピュータシステム及びコンピュータシステムの制御方法
WO2010124355A1 (en) 2009-05-01 2010-11-04 Nortel Networks Limited E-spring support of ethernet protection
CN102142976B (zh) * 2010-02-01 2015-04-01 中兴通讯股份有限公司 一种网状网保护域的共享保护方法及系统
CN101860481A (zh) * 2010-05-25 2010-10-13 北京邮电大学 一种MPLS-TP over OTN多层网络中区分优先级的业务传送方法及其装置
CN102035714B (zh) * 2010-12-17 2016-03-30 中兴通讯股份有限公司 一种相交相切环网保护方法及系统
CN102684973B (zh) * 2011-03-09 2017-03-29 中兴通讯股份有限公司 一种环网保护的方法及系统
CN102201985B (zh) * 2011-05-06 2014-02-05 杭州华三通信技术有限公司 多协议标签交换传送技术环保护倒换方法及节点
US9077641B2 (en) * 2011-05-15 2015-07-07 Orbit Communication Systems Ltd. Static ring network for vehicle communications
US9369341B2 (en) * 2011-05-15 2016-06-14 Orbit Communication Systems Ltd. Static ring network for vehicle communications
CN102387027B (zh) * 2011-06-25 2013-08-14 华为技术有限公司 一种网络配置方法、环形网络系统和一种节点
CN102307230B (zh) * 2011-08-18 2017-03-22 中兴通讯股份有限公司 业务传输方法和装置
CN102299835B (zh) 2011-08-26 2014-01-08 华为技术有限公司 环网故障切换方法和装置
WO2012167516A1 (zh) 2011-09-06 2012-12-13 华为技术有限公司 在环形拓扑上生成标签转发表的方法、装置和系统
WO2012167553A1 (zh) * 2011-11-15 2012-12-13 华为技术有限公司 确定tdm传输中时隙配置的方法和设备
CN102437957B (zh) * 2011-12-16 2015-07-08 华为技术有限公司 一种多协议标签交换的相交环处理方法及装置
CN102833109B (zh) * 2012-08-30 2015-01-07 华为技术有限公司 故障点位置信息处理方法及设备
CN102833108B (zh) * 2012-08-30 2016-03-30 华为技术有限公司 故障点位置信息处理方法及设备
CN102970219B (zh) * 2012-11-30 2016-03-30 华为技术有限公司 绑定保护环的方法和装置
CN103179035A (zh) * 2013-03-01 2013-06-26 苏州大学 一种网络固定路径选择方法、装置及光传输网络
US9125084B2 (en) 2013-03-14 2015-09-01 Aviat U.S., Inc. Systems and methods for performing layer one link aggregation over wireless links
US9509598B2 (en) * 2013-08-02 2016-11-29 Time Warner Cable Enterprises Llc Apparatus and methods for intelligent deployment of network infrastructure based on tunneling of ethernet ring protection
CN104639362A (zh) * 2013-11-15 2015-05-20 中兴通讯股份有限公司 Oam性能监控方法及装置
US9729455B2 (en) * 2014-06-30 2017-08-08 Juniper Networks, Inc. Multi-protocol label switching rings
US9692693B2 (en) 2014-06-30 2017-06-27 Juniper Networks, Inc. Bandwidth control for ring-based multi-protocol label switched paths
US10218611B2 (en) 2014-06-30 2019-02-26 Juniper Networks, Inc. Label distribution protocol (LDP) signaled multi-protocol label switching rings
US9973377B2 (en) 2014-12-19 2018-05-15 Futurewei Technologies, Inc. Protection against dual failures for resilient two layer networks
CN112087386B (zh) 2015-09-22 2024-01-02 华为技术有限公司 一种报文处理的方法、设备和系统
CN114915459B (zh) * 2015-12-17 2024-05-24 郭爱波 一种环形传输装置
JP2018527765A (ja) * 2016-08-03 2018-09-20 北京小米移動軟件有限公司Beijing Xiaomi Mobile Software Co.,Ltd. サービス接続を確立する方法、装置、プログラム、及び記録媒体
US11233748B1 (en) 2018-08-30 2022-01-25 Juniper Networks, Inc. Bandwidth management for resource reservation label switched path of a ring network
CN111294669B (zh) * 2018-12-07 2022-04-29 中国移动通信集团内蒙古有限公司 光传送网otn网络架构及调度方法
CN109547343B (zh) * 2019-01-04 2020-09-18 网宿科技股份有限公司 一种流量调度方法及系统
CN110557192B (zh) * 2019-07-22 2022-11-29 华信塞姆(成都)科技有限公司 一种基于FlexE的SPN中的双节点互连环保护方法
CN116349208A (zh) * 2020-09-25 2023-06-27 华为技术有限公司 路由方法及数据转发系统
CN112511611B (zh) * 2020-11-19 2021-08-10 腾讯科技(深圳)有限公司 节点集群的通信方法、装置、系统及电子设备
CN113787605B (zh) * 2021-09-24 2022-08-09 北新集团建材股份有限公司 一种石膏板生产线网络连接系统
CN114039859B (zh) * 2021-11-03 2023-05-30 中盈优创资讯科技有限公司 一种stn网络设备链改环方法及装置
CN114124782B (zh) * 2021-11-24 2023-02-28 北京鼎兴达信息科技股份有限公司 终端ip业务路径的确定方法
CN115002034B (zh) * 2022-06-16 2023-06-16 北京经纬恒润科技股份有限公司 一种报文通信网络构建、报文传输方法及装置
CN115225425B (zh) * 2022-09-19 2022-12-27 北京天辰合创科技有限公司 动态可插入环形网络架构数据采集方法及系统
CN116846946A (zh) * 2023-06-16 2023-10-03 泽拓科技(深圳)有限责任公司 合并网络连接的通信方法、系统、存储介质及设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2003050553A1 (en) * 2001-12-12 2003-06-19 Riverstone Networks, Inc. Method and system for providing failure protection in a ring network that utilizes label switching
JP2003224586A (ja) * 2002-01-29 2003-08-08 Nec Corp 二重リングネットワークにおける折り返しプロテクションのためのシグナリング方式
US20030227919A1 (en) * 2002-06-05 2003-12-11 Corrigent Systems Ltd. Signaling MPLS over RPR rings
CN1472938A (zh) * 2002-08-02 2004-02-04 ���Ͽع����޹�˾ 实现rpr网和mpls网之间端对端连接的方法和设备

Family Cites Families (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
ATE309658T1 (de) * 1992-07-20 2005-11-15 Siemens Ag Atm-kommunikationssystem
DE4332824C1 (de) * 1993-09-27 1995-03-16 Siemens Ag Verfahren und Schaltungsanordnung zum Übertragen von Nachrichtenzellen über virtuelle Pfade eines ATM-Kommunikationssystems
GB9601692D0 (en) * 1996-01-27 1996-03-27 Newbridge Networks Corp Network with ring architecture
US5838924A (en) * 1996-08-06 1998-11-17 Lucent Technologies Inc Asynchronous transfer mode (ATM) connection protection switching apparatus and method
US20030133417A1 (en) * 1997-03-12 2003-07-17 Sig H. Badt Method and message therefor of monitoring the spare capacity of a dra network
US5987516A (en) * 1997-10-07 1999-11-16 Nortel Networks Corporation Method of gathering information pertaining to calls between nodes in a connection-oriented network
JP3699837B2 (ja) * 1998-10-30 2005-09-28 株式会社東芝 ルータ装置及びラベルスイッチパス制御方法
US6356544B1 (en) * 1999-05-03 2002-03-12 Fujitsu Network Communications, Inc. SONET add/drop multiplexer with packet over SONET capability
US6594232B1 (en) * 1999-06-02 2003-07-15 Marconi Communications, Inc. Transmitter-based path protection switching in a ring network
US6532088B1 (en) * 1999-09-10 2003-03-11 Alcatel System and method for packet level distributed routing in fiber optic rings
JP2002198994A (ja) * 2000-12-26 2002-07-12 Nec Corp Gfpフレーム転送装置およびgfpフレーム転送方法
US20020110087A1 (en) * 2001-02-14 2002-08-15 David Zelig Efficient setup of label-switched connections
US20020126690A1 (en) * 2001-02-26 2002-09-12 Maple Optical Systems, Inc. Packet transmission scheduling in a data communication network
US7068654B1 (en) * 2001-04-18 2006-06-27 3Com Corporation System and method for providing masquerading using a multiprotocol label switching
JP4161557B2 (ja) * 2001-09-03 2008-10-08 株式会社日立製作所 パケット転送方法及びその装置
EP1294202A1 (en) * 2001-09-18 2003-03-19 Lucent Technologies Inc. A method of sending data packets through a MPLS network, and a MPLS network
US7289437B2 (en) * 2001-10-10 2007-10-30 Alcatel Lucent System and method for routing stability-based integrated traffic engineering for GMPLS optical networks
US7006440B2 (en) * 2001-10-26 2006-02-28 Luminous Networks, Inc. Aggregate fair queuing technique in a communications system using a class based queuing architecture
US20030208525A1 (en) * 2002-05-06 2003-11-06 Vinay Bannai System and method for providing transparent lan services
US7397761B2 (en) * 2002-06-10 2008-07-08 Lucent Technologies Inc. Routing restorable service-level-guaranteed connections using maximum 2-route flows
US7545735B1 (en) * 2003-03-11 2009-06-09 Atrica Israel Ltd. Scalable protection mechanism for hierarchical multicast service in ring based networks
US20040213264A1 (en) * 2003-04-25 2004-10-28 Nortel Networks Limited Service class and destination dominance traffic management
WO2004102853A2 (en) * 2003-05-06 2004-11-25 Overture Networks, Inc. Protected switching ring
US7366109B2 (en) * 2003-10-29 2008-04-29 Nortel Networks Limited Virtual private networks within a packet network having a mesh topology
CN100440843C (zh) * 2004-05-12 2008-12-03 华为技术有限公司 一种环网及其业务实现方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2003050553A1 (en) * 2001-12-12 2003-06-19 Riverstone Networks, Inc. Method and system for providing failure protection in a ring network that utilizes label switching
JP2003224586A (ja) * 2002-01-29 2003-08-08 Nec Corp 二重リングネットワークにおける折り返しプロテクションのためのシグナリング方式
US20030227919A1 (en) * 2002-06-05 2003-12-11 Corrigent Systems Ltd. Signaling MPLS over RPR rings
CN1472938A (zh) * 2002-08-02 2004-02-04 ���Ͽع����޹�˾ 实现rpr网和mpls网之间端对端连接的方法和设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP1796328A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113785280A (zh) * 2019-03-27 2021-12-10 图核有限公司 环形计算机网络上的嵌入环

Also Published As

Publication number Publication date
EP1796328A4 (en) 2007-11-21
US20070280251A1 (en) 2007-12-06
CN100334857C (zh) 2007-08-29
EP1796328A1 (en) 2007-06-13
CN1756222A (zh) 2006-04-05

Similar Documents

Publication Publication Date Title
WO2006034644A1 (fr) Reseau en anneau et methode de mise en oeuvre de ses services
EP1748599B1 (en) A ring carrying network and the method for implementing the service carry
EP1739914B1 (en) Method, apparatus, edge router and system for providing a guarantee of the quality of service (qos)
JP5106100B2 (ja) アドレス型キャリアネットワークにおける区別転送
US7872991B2 (en) Methods and systems for providing MPLS-based layer-2 virtual private network services
EP1585259B1 (en) System and method for providing a multiple-protocol crossconnect
US7978701B2 (en) Virtual ethernet MAC switching
WO2009035808A1 (en) Systems and methods for a self-healing carrier ethernet topology
EP1940083A2 (en) RPR transmission route designation method and apparatus
WO2007098663A1 (fr) Interface de liaison dans une boucle rpr, procédé de mise en oeuvre et appareil
WO2006039871A1 (fr) Procede pour la transmission de message de commande dans un reseau en boucle a commutation multiprotocole par etiquette
WO2006042470A1 (fr) Procede de transmission equitable de flux dans un reseau local en boucle mpls
WO2006042472A1 (fr) Procede de mise en oeuvre de transmission satisfaisante de flux dans le reseau en anneau mpls
Wang Optical Ethernet: Making Ethernet carrier class for professional services
Yongjun et al. Ring Network Technology for Carrier Ethernet
Van de Voorde et al. Carrier-grade Ethernet: extending Ethernet into next generation metro networks
Foisel et al. Evaluation of IP over WDM network architectures
Liu Ethernet transmisson solution for SONET/SDH system
Li et al. Enhanced metro MSTP and its applications
Aguirre-Torres et al. Packet-aware transport for video distribution
Berger Delivering Ethernet services in the Metro
Barenco Abbas et al. Scalable QoS approach in a core internet network
Chamas A truly end-to-end global multiservice optical ethernet networking architecture
Lian Design and analysis of optical metropolitan networks
Huynh et al. Metropolitan Ethernet Network

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KM KP KR KZ LC LK LR LS LT LU LV LY MA MD MG MK MN MW MX MZ NA NG NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU LV MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 2005792149

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

WWP Wipo information: published in national office

Ref document number: 2005792149

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 11579029

Country of ref document: US

WWP Wipo information: published in national office

Ref document number: 11579029

Country of ref document: US

WWW Wipo information: withdrawn in national office

Ref document number: 2005792149

Country of ref document: EP