CN116527571A - Cross-region wide area network two-layer transparent virtual circuit data transmission system - Google Patents

Abstract

The invention relates to a cross-regional wide area network two-layer transparent virtual circuit data transmission system, which comprises a plurality of POP devices deployed in different cities, wherein each POP device comprises 2 PE routers, and each PE router is recorded as a PE node; two PE nodes on the POP equipment are respectively connected with two PE nodes on the rest of the POP equipment; PE nodes among different cities form ring network protection. When any PE node or any link on the POP device fails, the network connection automatically switches to the remaining PE nodes and links to reach the corresponding POP device. Compared with the prior art, the method and the device can be used for rapidly switching to the backup route when one loop fails, and have the advantages of good fault tolerance, flow saving, high reliability and the like.

Description

Cross-region wide area network two-layer transparent virtual circuit data transmission system

Technical Field

The invention relates to the technical field of network communication, in particular to a cross-region wide area network two-layer transparent virtual circuit data transmission system.

Background

At present, the data private line access platform is almost the platform of the basic link of a single operator, namely, the interconnection among the cross-region wide area network platform nodes is mostly the single operator line, and the following problems can occur:

1. the problem of reliability and fault tolerance of a single operator bottom link can bring huge economic loss to clients if the current node or optical cable is excavated or broken, so that the platform equipment can not be used or recovered in a short time;

2. the existing platform has a backup scheme, when a major machine room has a major fault, the main machine room is manually switched to the backup machine room, the switching time limit of the scheme is slow, manual intervention is needed, and the full-automatic switching without packet loss cannot be realized;

3. the loop protection of the existing operators is single, the bottom optical fiber is generally self-built, the links of the operators are single, and the mixed networking of the links of a plurality of operators cannot be realized;

4. the conventional operator platform cannot realize the hybrid networking of multiple links, most of the links are networking by a single MSTP circuit mode, and the networking cannot be realized by utilizing a broadband IP network or an MPLS VPN mode.

5. The existing two-layer network architecture of the operator has a certain defect in design due to the existence of many years, is generally limited to the range of a metropolitan area network, and cannot be expanded to a two-layer transparent networking in a cross-region mode.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide a cross-region wide area network two-layer transparent virtual circuit data transmission system.

The aim of the invention can be achieved by the following technical scheme:

a cross-region wide area network two-layer transparent virtual circuit data transmission system comprises a plurality of POP devices, wherein each POP device comprises 2 PE routers, and each PE router is recorded as a PE node;

two PE nodes on the POP equipment are respectively connected with two PE nodes on the rest of the POP equipment, and multipath ring network protection is provided between the POP equipment;

each PE node interface is bound with MPLS LDP protocol;

when any PE node or any link on the POP device fails, the network connection automatically switches to the remaining PE nodes and links to reach the corresponding POP device.

Further, connected PE nodes on different POP devices are bound with a BFD bidirectional forwarding detection protocol. BFD is a lightweight detection protocol, and can realize millisecond detection of whether the bottom line fails, and when the failure is detected, the upper LDP protocol is immediately told, and the backup LSP channel is automatically switched.

Further, the PE routers on the POP equipment are routers adopting an OSPF protocol or an ISIS protocol.

Further, the PE router on the POP equipment is provided with an LDP-IGP synchronization mechanism;

LDP-IGP synchronization is a technique that ensures that LDP and OSPF/ISIS cooperate to minimize traffic loss time in the event of a network failure by synchronizing the states between LDP and OSPF/ISIS. When the main link fails, the LDP-IGP synchronization mechanism delays the back switching of the route by setting the cost value of the IGP until the LDP completes convergence, thereby preventing traffic loss.

Further, the PE router on the POP equipment is provided with an LDP FRR fast reroute mechanism. LDP FRR provides rapid rerouting function for MPLS backbone network, realizing link backup; when the main LSP fails, the traffic is quickly switched to the backup path, thereby avoiding the loss of the traffic to the greatest extent.

Further, the PE router on the POP device is provided with an LDP GR restarting mechanism. The LDP GR (Graceful Restart) uses the separation characteristic of the MPLS forwarding plane and the control plane to realize that the forwarding is not interrupted when the protocol is restarted or the main/standby is switched.

Further, a local adjacency and remote adjacency LDP session coexistence mechanism is arranged between two connected PE nodes.

Further, the PE routers on the POP equipment are routers supporting MPLS L2 VPN.

Further, the PE node is connected with a plurality of users, and the users are connected with the PE node through MSTP, PTN or bare optical fibers.

Further, the PE nodes between different POP devices are connected through MSTP, MPLS VPN or IPsec encrypted data channels.

Compared with the prior art, the invention has the following beneficial effects:

1. according to the invention, two PE routers are deployed on the POP equipment arranged in different cities, the PE routers in the POP equipment in different cities are connected with each other to form ring network protection, and when any PE node or any link on the POP equipment fails, the network connection is automatically switched to other PE nodes and links so as to reach the corresponding POP equipment, so that the ring network protection has good fault tolerance.

2. The invention delays the back-cut of the route by setting the cost value of the IGP by deploying the LDP-IGP synchronization mechanism when the main link fails, and prevents the flow loss during switching to the backup machine room when the main machine room fails greatly until the LDP completes convergence.

3. The invention has decoupling property of interconnection links and bottom links among all city nodes, can be used for combining networks through various links such as MSTP, bare optical fiber, SDH, broadband IP network, MPLS VPN and the like, improves the reliability of various links, reduces the networking cost, and provides a better and lower-cost two-layer transparent link solution.

4. The invention has protocol transparent transmission, realizes the two-layer transparent networking of providing a user LAN to LAN in a cross-region and cross-domain mode, does not process upper layer protocols of clients, and directly transmits the protocols in a transparent mode.

Drawings

FIG. 1 is a schematic diagram of a system architecture of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

Noun interpretation

OSPF protocol: open Shortest Path First, open shortest path first protocol, is a link state based interior gateway protocol IGP (Interior Gateway Protocol) developed by IETF organizations.

ISIS protocol: intermediate system-to-intermediate system IS-IS (Intermediate System to Intermediate System) belongs to interior gateway protocol IGP (Interior Gateway Protocol) for use within autonomous systems. IS-IS also a link state protocol that uses the shortest path first SPF (Shortest Path First) algorithm for route calculation.

MPLS protocol: multiprotocol label switching MPLS (Multiprotocol Label Switching) is a IP (Internet Protocol) backbone technology. MPLS introduces a connection-oriented label switching concept on a connectionless IP network, and combines a third layer routing technology and a second layer switching technology, thereby fully playing the flexibility of IP routing and the simplicity of two layer switching.

LDP protocol: the label distribution protocol LDP (Label Distribution Protocol) is a control protocol of multiprotocol label switching MPLS, and corresponds to a signaling protocol in a conventional network, and is responsible for operations such as classification of forwarding equivalence classes FEC (Forwarding Equivalence Class), allocation of labels, and establishment and maintenance of a label switching path LSP (Label Switched Path). LDP specifies various messages and related processes in the label distribution process.

BFD protocol: the bidirectional forwarding detection BFD (Bidirectional Forwarding Detection) is a unified detection mechanism of the whole network, and is used for rapidly detecting and monitoring forwarding connectivity status of links or IP routes in the network.

LDP FRR protocol: LDP FRR (Fast Reroute) provides rapid rerouting function for MPLS network, realizing link backup; when the main LSP fails, the traffic is quickly switched to the backup path, thereby avoiding the loss of the traffic to the greatest extent.

As shown in FIG. 1, the scheme provided by the invention is illustrated by four cities of Beijing, shanghai, guangzhou and Shenzhen, wherein SH-PE1/SH-PE2 represents PE1 and PE2 nodes of Shanghai PoP equipment;

BJ-PE1/BJ-PE2 represents PE1 and PE2 nodes of Beijing PoP equipment;

GZ-PE1/GZ-PE2 PE1 and PE2 nodes representing Guangzhou PoP devices;

SZ-PE1/SZ-PE2 represents PE1 and PE2 nodes of Shenzhen PoP equipment;

USER1/USER2/USER 3..usern represents a group of USERs connected point-to-point.

The specific scheme adopted by the embodiment comprises the following steps:

1. PE equipment of North Shang Sheng PoP equipment is hosted in a local IDC data center machine room, and environment and electric facilities are provided by the machine room, so that the safety guarantee problem of the equipment is solved;

each POP device adopts a high-end router in 2 PE, and each city can adopt MSTP, MPLS VPN and IPsec encrypted data channels of three major operators to connect PE nodes of each city, so as to form the eight diagrams topology structure diagram shown in figure 1, and realize the following functions:

loop protection exists between 4 PE nodes in 1.1 or any 2 cities, such as a looped network is formed between Beijing and Shanghai, a looped network is formed between Beijing and Guangzhou, a looped network is formed between Beijing and Shenzhen, and a looped network is formed between Shenzhen and Guangzhou.

1.2, and any 3 cities form a medium double-loop protection among 6 nodes, such as Beijing, shanghai and Guangzhou PE nodes forming a double-loop protection;

a small ring network protection is formed between 4 nodes of 1.3 and 4 cities;

1.4, 4 cities and 8 nodes form a large ring network protection;

all cities are protected by multipath ring networks, and when any one node fails, any link can be automatically switched to other nodes and links to reach the corresponding city.

2. IGP routing and LDP reliability design scheme

2.1, IGP route adopts OSPF dynamic route, add every PE node interface into OSPF area0 area, interface binding MPLS protocol and LDP label distributing protocol, far end interface definition OSPF spending is 10, and use GE port direct connection with PoP equipment 2 PE, c in this embodiment the ost value is defined as follows: the Shanghai PE1 and PE2 are defined as a cost1, the Beijing PE1 and PE2 are defined as a cost2, the Guangzhou PE1 and PE2 are defined as a cost3, and the Shenzhen PE1 and PE2 are defined as a cost4.

2.2, the ospf protocol and the LDP label issuing protocol of each interface bind BFD bidirectional forwarding detection protocol, BFD is a lightweight detection protocol, can realize millisecond detection of whether the bottom line breaks down, and immediately tell the upper LDP protocol to automatically switch to a backup LSP channel when the break down is detected.

2.3, BFD binding LDP LSP implementation process is as follows:

BFD binds an LDP LSP, which refers to establishing a BFD session on the LSP link and binding the session with the LSP. And the BFD is utilized to rapidly detect the failure of the LSP link and trigger the flow switching of the LSP. When BFD is used to detect a unidirectional LSP path, the reverse link may be an IP link, LSP tunnel.

When detecting connectivity of an LDP LSP, the BFD session uses a static negotiation approach: the session is established by the negotiation mechanism of the BFD itself by manually configuring the local identifier and the remote identifier of the BFD. The user needs to specify the next hop IP address of the LSP to be detected and the opposite address of the BFD session to which the LSP will be bound.

The BFD will then use an asynchronous mode to detect the connectivity of the LSP, i.e. the BFD messages are sent periodically to each other between Ingress and Egress. If any one end does not receive the BFD message sent by the opposite end in the detection time, the LSP is considered to have faults, and LSP fault events are reported to the LDP management module.

2.4, LDP and OSPF/ISIS synchronous linkage design scheme

LDP-IGP synchronization is a technique that ensures that LDP and OSPF/ISIS cooperate to minimize traffic loss time in the event of a network failure by synchronizing the states between LDP and OSPF/ISIS.

In a networking in which a primary link exists, when the primary link fails, both OSPF/ISIS and LSP switch to the backup link. However, when the primary link recovers from the failure, since the OSPF/ISIS converges faster than the LDP, the OSPF/ISIS switches back to the primary link before the LDP, and at this time, the LSP of the primary link cannot be established immediately, and some time is required for preparation before establishment, such as adjacency recovery, etc., thus resulting in loss of the LSP traffic. When an LDP session or adjacency between the main link nodes fails, the LSP of the main link is deleted, but the OSPF/ISIS still uses the main link, resulting in that LSP traffic cannot be switched to the backup link and traffic is continuously lost.

The basic principle of LDP and OSPF/ISIS synchronization is: the back-cut of the route is deferred by setting the cost value of the OSPF/ISIS until the LDP completes convergence. That is, before the LSP of the main link is established, the backup link LSP is reserved, so that the traffic is continuously forwarded from the backup link. And deleting the backup LSP until the LSP of the main link is successfully established.

2.5 automatic binding FRR rerouting function of LDP protocol

The automatic LDP FRR provides a rapid rerouting function for the MPLS backbone network, so that the link backup is realized; when the main LSP fails, the traffic is quickly switched to the backup path, thereby avoiding the loss of the traffic to the greatest extent.

In MPLS backbone networks, when the main link fails, although IP FRR makes OSPF/ISIS routes converge quickly and switch to backup paths, LSPs are re-established for MPLS networks, and this process cannot avoid loss of traffic. In addition, when the LSP fails (caused by a non-primary link failure), traffic forwarding can only be resumed after the LSP is re-established, which can cause long disruption of MPLS traffic. There is therefore a need for a solution that can provide fast reroute in an MPLS network, i.e. an automated LDP FRR scheme.

The automatic LDP FRR firstly acquires a Liberal Label through a free Label holding mode (Liberal) of LDP signaling, applies for forwarding list item resources for the Label, and issues forwarding information to a forwarding plane as a standby forwarding list item of a main LSP. When the interface fails (the interface senses itself or combines with BFD detection) or the main LSP is not enabled (combines with BFD detection), the flow can be quickly switched to the backup path, thereby realizing the protection of the main LSP.

Automated LDP FRR: depending on the implementation of IP FRR. Only if the source of the Liberal Label matches the existing backup route, that is, the reserved Liberal Label comes from the backup route outgoing interface and the next hop, and the backup LSP triggering strategy is satisfied, the backup LSP can be established for the backup route and the forwarding table item can be issued.

2.6, LDP Graceful Restart design

The LDP GR (Graceful Restart) uses the separation characteristic of the MPLS forwarding plane and the control plane to realize that the forwarding is not interrupted when the protocol is restarted or the main/standby is switched.

In an MPLS network, when a device protocol is restarted or the active-standby is switched, a device deletes a label forwarding table entry on a forwarding plane, resulting in interruption of data forwarding.

This problem can be solved by LDP GR, improving the reliability of the network. When the LDP GR is restarted or the main/standby switching is performed, the label forwarding list item is reserved by utilizing the characteristic that a control plane and a forwarding plane are separated, and the equipment still forwards the message according to the list item, so that the data transmission is ensured not to be interrupted. Meanwhile, after the protocol is restarted or the main/standby is switched, the equipment is restored to the state before restarting under the assistance of the neighbor.

2.7 local and remote LDP Session coexistence design

The present remote LDP session co-exists, which is essentially that the local and remote adjacencies of LDP can be bound to the same peer at the same time, i.e. the peer is maintained by both local and remote adjacencies at the same time.

When a link associated with a local adjacency fails, resulting in the adjacency being deleted, the peer type can only be changed (the peer type is determined by the type of adjacency maintaining the peer, switching can be made between local, remote, and local remote coexistence) without affecting the presence and status of the peer.

During a link failure or failure recovery, the peer type may change, as may the session type corresponding to the peer. However, the session is not deleted or Down set in the whole process, and can always function.

The specific implementation process is as follows:

taking PE nodes of Shanghai and Beijing as examples, the two PE nodes of Shanghai are respectively denoted as SH-PE1 and SH-PE2, and the two PE nodes of Beijing are respectively denoted as BJ-PE1 and BJ-PE 2;

taking SH-PE1 and BJ-PE1 as examples to describe the whole process proposed by the present invention, MPLS LDP is enabled on two routes SH-PE1 and BJ-PE1 that are directly connected, respectively, to establish a local session between SH-PE1 and BJ-PE 1. Then, the SH-PE1 and the BJ-PE1 are configured to be remote peers, and a remote session between the SH-PE1 and the BJ-PE1 is established. In this case, SH-PE1 and BJ-PE1 are both local and remote neighbors, and the local and remote sessions between SH-PE1 and BJ-PE1 coexist. L2VPN signaling messages are communicated through this session.

The physical link between SH-PE1 and BJ-PE1 goes into the Down state, causing the local adjacency of the LDP peer to go into the Down state. The routing between SH-PE1 and BJ-PE1 is achieved through P, namely the far-end adjacent body is still in the Up state, the session type is changed to become a far-end session, but the session is still in the Up state, the L2VPN senses no change of the session state and cannot be actively dismantled, the process that the L2VPN disconnects the adjacent body and resumes is avoided, and the service interruption time is reduced.

The failure then resumes, the link between SH-PE1 and BJ-PE1 goes into the Up state, and the local adjacency then goes into the Up state. The session type changes and is restored to the far coexistence session, the session is still in the Up state, the L2VPN cannot sense the change of the session state and cannot be actively dismantled, and therefore service interruption time is reduced.

3. User access network design scheme

The user office point or the data center is accessed to the PE node of the urban PoP equipment by the mode of MSTP, PTN, bare optical fiber and the like of a local operator, and double routing or single routing can be realized. Both end ACs can bind a VC ID through MPLS L2VC to isolate each user, so as to ensure the security of two layers. A USER1 USER in the network structure diagram is respectively connected to Shanghai SH-PE1 and Shenzhen SH-PE1 to form a VC virtual circuit connection, and the upper layer of the virtual circuit can bear various protocols and application layer data message transmission.

In summary, the invention provides an innovative network topology architecture, poP points between any 2 cities have loop protection, loop protection is formed between any 4 nodes of 4 cities, such as a ring network between Beijing and Shanghai, a ring network between Beijing and Guangzhou, a ring network between Beijing and Shenzhen, a ring network between Shenzhen and Guangzhou, a medium ring network is formed between any 6 nodes of 4 cities, a large ring network is formed between 8 nodes of 4 cities, multipath ring network protection is provided between all cities, and when any one node fails, any link can be automatically switched to other nodes and links reach the corresponding city.

In the scheme provided by the invention, the switching of the main loop and the standby loop of the node and the link can reach the millisecond level, and when one node or one link of any loop fails, the automatic switching of the 50ms level without packet loss to the standby route can be realized.

The invention realizes decoupling of the interconnection links and the bottom links among the PoP nodes in each city, can realize hybrid networking of various links such as MSTP, bare optical fiber, SDH, broadband IP network, MPLS VPN and the like, improves the reliability of various links, reduces networking cost, and provides a better, better and cheaper two-layer transparent link solution for customers;

the invention also has protocol transparent transmission, realizes the two-layer transparent networking of providing the user LAN to LAN in a cross-region and cross-domain mode, does not process the upper layer protocol of the client, and directly transmits the protocol in a transparent mode.

While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.

Claims (10)

1. The data transmission system of the two-layer transparent virtual circuit of the cross-regional wide area network is characterized by comprising a plurality of POP devices, wherein each POP device comprises 2 PE routers, and each PE router is recorded as a PE node;

two PE nodes on the POP equipment are respectively connected with two PE nodes on the rest of the POP equipment;

each PE node interface is bound with MPLS LDP protocol;

when any PE node or any link on the POP device fails, the network connection automatically switches to the remaining PE nodes and links to reach the corresponding POP device.

2. The system of claim 1, wherein the connected PE nodes on different POP devices are bound with BFD bidirectional forwarding detection protocol.

3. The system for transmitting data over a two-layer transparent virtual circuit of a regional wide area network of claim 1, wherein the PE routers on the POP device are routers using OSPF protocol or ISIS protocol.

4. The data transmission system of a two-layer transparent virtual circuit of a cross-regional wide area network according to claim 1, wherein a PE router on the POP device is provided with an LDP-IGP synchronization mechanism;

when the main link fails, the LDP-IGP synchronization mechanism delays the back switching of the route by setting the cost value of the IGP until the LDP completes convergence, thereby preventing traffic loss.

5. The system for transmitting data over a two-layer transparent virtual circuit of a regional wide area network of claim 1, wherein the PE router on the POP device is provided with an LDP FRR fast reroute mechanism.

6. The system for transmitting data over a two-layer transparent virtual circuit of a regional wide area network of claim 1, wherein the PE router on the POP device is provided with an LDP GR restart mechanism.

7. The system for data transmission over a two-layer transparent virtual circuit in a regional wide area network of claim 1, wherein a local adjacency and remote adjacency LDP session coexistence mechanism is provided between two connected PE nodes.

8. The system for transmitting data over a two-layer transparent virtual circuit of a regional wide area network according to claim 1, wherein the PE routers on the POP device are routers supporting MPLS L2 VPN.

9. The system for transmitting data over a two-layer transparent virtual circuit of a regional wide area network according to claim 1, wherein a plurality of subscribers are connected to the PE node, and the subscribers are connected to the PE node by MSTP, PTN or bare optical fibers.

10. The system of claim 1, wherein the PE nodes between different POP devices are connected by MSTP, MPLS VPN, or IPsec encrypted data tunnels.