CN101931587A - Virtual cluster routing method and system - Google Patents

Abstract

The embodiment of the present invention relates to a virtual cluster routing method and system, and the method is applicable to a virtual cluster routing system. The virtual cluster routing system includes a plurality of node devices interconnected through ports; A packet of identification information; search the forwarding information base FIB according to the target identification information, and obtain the outgoing interface identification of this node, and the corresponding relationship between the purpose identification information and the outgoing interface identification of this node is included in the FIB; The outgoing interface corresponding to the interface identifier sends the packet to the adjacent node device connected to the outgoing interface. The embodiment of the present invention aggregates destination identification information and outgoing interface identifications of multiple interconnected router devices into a FIB based on software technology, so that the multiple interconnected router devices are aggregated into a virtual cluster routing system.

Description

Virtual cluster routing method and system

technical field

The invention relates to the communication field, in particular to a virtual cluster routing method and system.

Background technique

With the continuous development of communication technology, the application scale of cluster routers in the core network and super nodes has become unstoppable.

A cluster router is also called a multi-chassis router. It uses multi-chassis hardware cluster technology (that is, multi-level full switching network technology) to connect multiple high-end routers in a certain way to form a multi-level and multi-plane switching matrix system. Make it work together, and only appear as a logical router to the outside world, so as to break through the limitations of a single chassis in terms of switching capacity, power consumption, and heat dissipation, and smoothly expand to a routing and switching system with larger capacity. For the existing cluster routers, in order to adapt to the huge data exchange capacity between the devices in the cluster, multiple high-end routers are connected to the central switch matrix chassis through ultra-short-distance optical fibers, forming a four-for-one, eight for two, or ten for four Sixth and other large cluster routers.

In the existing cluster routers, due to the introduction of a dedicated central switch matrix machine and ultra-short-distance optical fiber interconnection technology, the network topology is complex, and the routing convergence and stability are poor.

Contents of the invention

The embodiment of the present invention provides a virtual cluster routing method and system. Based on software technology, the control and management planes of multiple router devices interconnected through ports are aggregated into a control and management plane of a virtual cluster routing system, and the control and management plane of each device is uniformly controlled. The forwarding information base FIB makes the virtual cluster routing system appear as a routing node to the outside, thereby simplifying the network topology, simplifying operation and maintenance management, and improving routing convergence and stability.

An embodiment of the present invention provides a virtual cluster routing method, which is suitable for a virtual cluster routing system, and the virtual cluster routing system includes a plurality of node devices interconnected through ports; including:

The node device receives a message carrying destination identification information;

Searching the forwarding information base FIB according to the target identification information to obtain the outgoing interface identification of this node, the FIB includes the corresponding relationship between the purpose identification information and the outgoing interface identification of this node;

Sending the message to the adjacent node device connected to the outgoing interface through the outgoing interface corresponding to the outgoing interface identifier of the local node.

The embodiment of the present invention also provides a virtual cluster routing system, including: the virtual cluster routing system includes: a plurality of node devices interconnected through ports, and the node devices also include:

A receiving module, configured to receive a message carrying purpose identification information;

A search module, configured to search the forwarding information base FIB according to the target identification information, and obtain the outgoing interface identification of the current node, and the FIB includes the corresponding relationship between the target identification information and the outgoing interface identification of the current node;

A sending module, configured to send the message to an adjacent node device connected to the outgoing interface through the outgoing interface corresponding to the outgoing interface identifier of the local node.

It can be seen from the above technical solution that the embodiment of the present invention aggregates the destination identification information and the outgoing interface identification of multiple interconnected router devices into a FIB based on software technology, so that the multiple interconnected router devices are aggregated into a virtual cluster routing system , when forwarding the message, look up the FIB according to the destination identification information carried in the message, obtain the outbound interface ID of the node, and send the message through the outbound interface corresponding to the outbound interface ID of the node, so that the entire virtual cluster routing system It is represented as a routing system, which simplifies the network topology, simplifies operation and maintenance management, and improves routing convergence and stability.

Description of drawings

FIG. 1a is a schematic flowchart of a virtual cluster routing method provided by Embodiment 1 of the present invention;

Figure 1b is a schematic structural diagram of the VNE in the embodiment of the present invention;

Figure 1c is a schematic structural diagram of a VNE in an embodiment of the present invention;

FIG. 2 is a schematic flowchart of a virtual cluster routing method provided in Embodiment 2 of the present invention;

FIG. 3 is a schematic flowchart of a virtual cluster routing method provided by Embodiment 3 of the present invention;

FIG. 4 is a schematic flowchart of a virtual cluster routing method provided in Embodiment 4 of the present invention;

FIG. 5 is a schematic flowchart of a virtual cluster routing method provided in Embodiment 5 of the present invention;

FIG. 6 is a schematic structural diagram of a virtual cluster routing system provided by Embodiment 6 of the present invention;

FIG. 7 is a schematic structural diagram of a virtual cluster routing system provided by Embodiment 7 of the present invention;

FIG. 8 is a schematic structural diagram of a virtual cluster routing system provided by Embodiment 8 of the present invention;

FIG. 9 is a schematic structural diagram of a virtual cluster routing system provided by Embodiment 9 of the present invention.

Detailed ways

Specific embodiments of the present invention will be described in further detail below in conjunction with the accompanying drawings.

FIG. 1a is a schematic flowchart of a virtual cluster routing method provided by Embodiment 1 of the present invention. As shown in Figure 1a, this embodiment may include the following steps:

It should be noted that the virtual cluster routing method provided in this embodiment is applicable to a virtual cluster routing system, and the virtual cluster routing system includes multiple node devices interconnected through ports.

Step 101, the node device receives a message carrying destination identification information;

Step 102, search the forwarding information base (Forwarding Information Base, hereinafter referred to as: FIB) according to the target identification information, and obtain the outgoing interface identification of this node, and the corresponding relationship between the target identification information and the outgoing interface identification of this node is included in the FIB;

Step 103: Send the message to the adjacent node device connected to the outgoing interface through the outgoing interface corresponding to the outgoing interface identifier of the local node.

The virtual cluster routing method provided in this embodiment can be applied to any grouping network with a hierarchical topology, for example: a traditional Point of Presence (hereinafter referred to as: PoP) network, a hierarchical aggregation network, etc., this embodiment The hierarchical aggregation network is used as an example for illustration, but it is not limited to the hierarchical aggregation network. The virtual cluster routing method provided in this embodiment is based on software technology to virtualize and aggregate multiple distributed node devices (ie common routers) interconnected through ports such as GE, POS, WDM, etc. in a two-layer or multi-layer aggregation network into one Or multiple virtual cluster routing systems, the overall behavior of each virtual cluster routing system is externally represented as a routing system. In this embodiment, a virtual cluster routing system is named as a virtual network element (Virtual Network Element, VNE for short) for illustration, but it is not limited to VNE. Among them, multiple node devices inside the VNE can be connected through connection technologies such as Ethernet (Ethernet), IP network, special networking technology (Special Networking, hereinafter referred to as SN) or multi-protocol label switching (MPLS). The specific connection technology You can choose according to actual needs.

Inside the VNE, the topological structure formed by multiple node devices can be in various forms such as star, ring, mesh or hybrid. In this embodiment, the topology structure of multiple node devices inside the VNE is described as a star topology, but is not limited to the star topology.

In this embodiment, the port through which the VNE is interconnected with the external network is called the VNE external port (External Port, hereinafter referred to as: ExtP). The port that interconnects the devices of each node inside the VNE is called the VNE internal port (Internal Port, hereinafter referred to as: IntP). A node device that provides a VNE external connection port in the VNE is called a VNE edge node (edge Node, hereinafter referred to as: eNode) device. A node device in a VNE that does not provide a VNE external connection port is called a VNE intermediate node (intermediate Node, hereinafter referred to as: iNode) device. There may be no intermediate node device inside the VNE, or multiple intermediate node devices. Among them, the node devices connected to the node devices inside a VNE at the same time are collectively referred to as the adjacent nodes of the node devices inside the VNE. If the adjacent nodes are inside the VNE, they are called adjacent nodes inside the VNE; if the adjacent nodes are in the VNE The external network side is called the adjacent node of the network side.

It should be noted that in the existing cluster routers, the routing devices in the cluster are connected based on the physical layer, and the internal interconnection is realized through the ultra-short-distance optical fiber interconnection technology, which is costly and consumes high energy. However, in this embodiment, by integrating the IP addresses and identification information corresponding to the ports of multiple node devices connected to each other into a unified FIB, based on the IP layer interconnection, no matter what type or model the node devices inside the VNE are , as long as the FIB contains the IP address and identification information corresponding to the port of the node device, the node device is considered to be a node device inside the VNE. Compared with the prior art, the present invention establishes the FIB by means of software, realizes the aggregation of multiple node devices in a heterogeneous manner, increases the capacity of the cluster router, and improves the expansibility of the cluster router.

Fig. 1b is a schematic structural diagram of a VNE in an embodiment of the present invention. As shown in Figure 1b, this embodiment aggregates the control and management planes of multiple router devices interconnected through ports into a VNE control and management plane based on software technology, and uniformly controls the forwarding plane of each node device in the VNE, so that The entire VNE appears as a routing system externally. Therefore, inside the VNE, the control and management plane of one or more node devices with strong performance and good scalability can be set as the master node (master Node, hereinafter referred to as: mNode), which is responsible for coordinating the routing calculation of the entire VNE, Service control and cluster management, and accept the connection and configuration management of the network management equipment on behalf of the VNE; the master node can be served by an edge node device or an intermediate node device, or an independent control and management device. In this embodiment, a high-performance The edge node device serves as the master node for illustration, but not limited thereto. The control and management planes of other node devices are set as slave nodes (slave Node, hereinafter referred to as: sNode), which accept the coordination and management of the master node. Fig. 1c is a schematic structural diagram of a VNE in an embodiment of the present invention. As shown in Figure 1c, when two or more node devices are set to act as master nodes in the VNE at the same time, the multiple master nodes are mutual backups, which can realize master-standby redundancy or load sharing. Among them, the master-slave redundancy and load sharing will be described in detail in subsequent embodiments, and will not be repeated here. It should be noted that a device serving as a slave node can be directly connected to a device serving as a master node, or it can be connected to a device serving as a master node through multiple Connect to the device acting as the master node after the hop.

Each node device in the VNE is set with a unique node ID to distinguish different node devices. Each port on the node device is set with a unique port identifier to distinguish the outgoing interface of the node on different node devices. For example, if the port marked as 21 in the slave node device is an external port, the identification of the external port can be It is expressed as ExtP21; if the port marked as 21 in the slave node device is an internal connection port, then the identification of the internal connection port can be represented as IntP21. According to different communication protocols, an address information can be assigned to each port, so that each connection The port can implement routing in the communication network, for example:

1. When the VNE is connected to each node device through Ethernet, the internal interconnection of the VNE adopts the Ethernet switching mechanism to realize internal routing, and each connection port of each node device is assigned a unique Media Access Control (Media Access Control, hereinafter Abbreviation: MAC) address, VNE completes internal routing according to the Mac address.

2. When the VNE internally connects each node device through the IP network, the internal interconnection of the VNE adopts the IP routing mechanism to realize internal routing. Each connection port of each node device is assigned a unique IP address, and the VNE completes internal routing based on the IP address .

3. When the VNE is connected to each node device through the MPLS network, the internal interconnection of the VNE uses the MPLS label switching mechanism to realize internal routing. Each connection port of each node device is assigned a unique MPLS label, and the VNE completes internal routing according to the MPLS label. .

In this embodiment, the description is made by taking the connection of each node device inside the VNE through an IP network as an example, but not limited to the IP network. Correspondingly, each connection port of the node device in this example is assigned a unique IP address.

In this embodiment, the node device inside the VNE stores a unified external forwarding information base (external Forwarding Information Base, hereinafter referred to as: eFIB) table, and each eFIB table stores multiple eFIB entries, and the eFIB entries perform external routing for the VNE basis, including the correspondence between the destination IP address and the system outbound interface identifier. As shown in Table 1, the eFIB includes the destination IP address, next-hop IP address, and system outbound interface identifier corresponding to each external port of the VNE.

Table 1 eFIB table stored by the master node device

Destination IP address next hop address System Outgoing Interface Identifier 120.0/16 12.0.0.2 ExtP11 128.0/16 128.0.0.2 ExtP21 129.0/16 129.0.0.2 ExtP31 130.0/16 130.0.0.2 ExtP41

Further, the eFIB table is generated by the node devices inside the VNE, and stored in each node device inside the VNE, and each node device stores the same eFIB table. Wherein, the manners for the node device inside the VNE to generate the eFIB table may include the following three ways, but are not limited to the following three ways.

The first is the distributed dynamic generation method, which specifically includes the following steps:

Step 111, the edge node inside the VNE receives the neighbor information sent by the network side.

Among them, the neighbor information is the routing information regularly sent to the VNE by adjacent network elements in the external network of the VNE. The neighbor information includes topology information on the network side, and may specifically include identification information and IP addresses of adjacent network elements in the external network. and other information.

The edge node devices inside the VNE can receive neighbor information sent by neighboring network elements in the external network.

Step 112, the neighbor information received by each node device inside the VNE is synchronized.

After the edge node of the VNE receives the neighbor information on the network side, it forwards it inside the VNE through internal routing, so that all node devices inside the VNE can receive the same neighbor information and realize the synchronization of the neighbor information inside the VNE.

Step 113 , each node device inside the VNE generates its own eFIB table according to the synchronized neighbor information.

The node device inside the VNE calculates the routing table of the VNE according to the synchronized neighbor information, and calculates the eFIB table, which contains multiple eFIB entries.

Since the synchronized neighbor information in each node device inside the VNE is the same, the eFIB table saved in each node device is also the same.

Further, in order to ensure that the eFIB tables stored in each node device are the same, each node device may also synchronize the generated eFIB tables after generating their respective eFIB tables.

The second method is centralized dynamic generation, which specifically includes the following steps:

Step 121, the master node device inside the VNE receives the neighbor information sent by the network side and/or other node devices.

In this step, a master node device is set inside the VNE. If the master node device is configured with a system outbound interface, the node device can receive the neighbor information sent by the network side and the neighbors reported by each edge node inside the VNE. Information, so that the master node device can aggregate the neighbor information received by all the edge nodes inside the VNE; if the master node device is an intermediate node device, the master node device can receive the neighbor information reported by each edge node inside the VNE.

Step 122, the master node device generates an eFIB table according to the neighbor information received by itself.

When the master node device calculates the eFIB table according to the received neighbor information, the eFIB table includes multiple eFIB entries.

Step 123, the master node device returns the generated eFIB table to the slave node device.

The master node device sends the generated eFIB table to all slave node devices inside the VNE through internal routing, so that each node device inside the VNE saves the same eFIB table.

The third is the external configuration generation method, specifically: the system administrator sets the eFIB table of each node device inside the VNE through the network management device; or, the slave node device inside the VNE receives the eFIB table sent by other node devices.

It should be noted that, in this embodiment, the node devices in the VNE can generate the eFIB table using the above three methods, and in the actual application process, the VNE uses the above three methods to update the eFIB table.

In this embodiment, the node devices inside the VNE also store their own internal forwarding information bases (internal Forwarding Information Base, hereinafter referred to as: iFIB), and each iFIB table stores a plurality of iFIB entries, and the iFIB is used for internal routing of the VNE. in accordance with. As described in Table 2, the iFIB contains the corresponding relationship between the system outbound interface ID and the local node outbound interface ID.

Table 2 iFIB table stored in master node device 1

Further, the above-mentioned iFIB table is generated by the node devices inside the VNE, and is stored in each node device inside the VNE, and each node device stores its own iFIB table. Wherein, the way for the node device inside the VNE to generate the iFIB table may include the following two ways, but is not limited to the following two ways.

The first one is a static configuration method. The static configuration method is specifically: the network administrator pre-sets the iFIB table of each node device inside the VNE through the network management device according to the network topology planning, and the iFIB table contains multiple iFIB entries.

Further, the static configuration method may also be: each node device inside the VNE receives the iFIB table sent by other node devices, and sets its own iFIB table according to the received iFIB table.

The second is the dynamic generation method.

When the iFIB table is generated dynamically, each node device in the VNE needs to support the dynamic neighbor link discovery function and be able to dynamically discover the topology inside the VNE. The dynamic generation method also includes two methods of distribution or concentration.

Among them, the distributed method is that each node device calculates its own internal routing table and saves the generated internal routing table as an iFIB table; the centralized method is that the master node device in the VNE calculates the internal routing table of each node device and saves the generated internal routing table The routing table is delivered to the slave node device, and the slave node device saves the internal routing table as its own iFIB table.

It should be noted that each node device in the VNE can be dynamically increased or decreased. When the node device is increased or decreased, the VNE uses the above two methods to regenerate the iFIB table of each node device.

Furthermore, both the eFIB table and the iFIB table in the same node device contain the system outbound interface identifier, so the eFIB table and the iFIB table in the same node device can be combined into a forwarding information base FIB table. As shown in Table 3: the FIB table contains multiple FIB entries, and the FIB is the basis for the VNE to route the received message, specifically including the correspondence between the destination identification information and the outgoing interface identification of the node.

Table 3 FIB table stored by the master node device

Destination IP address next hop address System Outgoing Interface Identifier The outbound interface ID of this node 120.0/16 12.0.0.2 ExtP11 ExtP11 128.0/16 128.0.0.2 ExtP21 IntP11 129.0/16 129.0.0.2 ExtP31 IntP12 130.0/16 130.0.0.2 ExtP41 IntP13

When the network side needs to route the message to the destination IP address such as: 130.0.0.10 through the virtual cluster routing method provided by this embodiment, the network side sends the message to the VNE, and the message carries destination identification information. The identification information can be the destination IP address, or an internal routing label. The internal routing label is encapsulated into the message for the node device inside the VNE to realize the forwarding of the message inside the VNE. The internal routing label contains the system outbound interface Identification and quality of service (Quality of service, hereinafter referred to as: Qos) type information, etc. The destination identification information in this embodiment is described by taking the destination IP address as an example, but is not limited to the destination IP address. When the node device in the VNE receives the message sent by the network side, the node device finds the corresponding FIB from the FIB table stored in itself (as shown in Table 3) according to the destination IP address: 130.0.0.10, the FIB contains the outbound interface ID of this node: IntP13. Afterwards, the node device sends the message to the adjacent node connected to the outgoing interface of the local node.

In this embodiment, since the outgoing interface identifier of the current node is different from the outgoing interface identifier of the system, the port through which the node device sends the message is an internal connection port, and the node device sends the message to an adjacent node through the internal connection port, and the The adjacent node repeats the above operations, and sends the message to the device corresponding to the destination IP address on the network side. When the outgoing interface identifier of the local node is the same as the outgoing interface identifier of the system, the port through which the node device sends the message is an external connection port, and the node device sends the message to an adjacent node through the external connection port.

In this embodiment, based on software technology, the destination identifier information and outgoing interface identifiers of multiple interconnected router devices are aggregated into FIB, so that the multiple interconnected router devices are aggregated into a virtual cluster routing system, and the packets are processed When forwarding, search the FIB according to the destination identification information carried in the message, obtain the outbound interface ID of the node, and send the message through the outbound interface corresponding to the outbound interface ID of the node, so that the entire virtual cluster routing system appears as a routing system externally, This simplifies the network topology, simplifies operation and maintenance management, and improves routing convergence and stability.

Further, in this embodiment, the node device inside the VNE merges the eFIB table and the iFIB table into the same FIB table. When the node device searches for the outgoing interface identifier of the node according to the destination IP address, it only needs to perform one search action to obtain the destination IP address. The outbound interface identifier of the node corresponding to the address improves efficiency.

FIG. 2 is a schematic flowchart of a virtual cluster routing method provided by Embodiment 2 of the present invention. This embodiment is based on the foregoing Embodiment 1, and describes a scenario where a node eFIB table and an iFIB table are separated. As shown in Figure 2, this embodiment may include the following steps:

It should be noted that the difference between this embodiment and the first embodiment is that the eFIB table and the iFIB table are separated from each other in this embodiment, which can reduce the complexity of the FIB table. Since the eFIB table and the iFIB table are separated from each other in this embodiment, the node device inside the VNE can also obtain the outbound interface identifier of the node corresponding to the destination IP address by performing two searches.

Before the steps of this embodiment are executed, the node devices inside the VNE need to generate the eFIB table and the iFIB table. The method for generating or updating the eFIB table and the iFIB table can refer to the description in the first embodiment above, and will not be repeated here.

Step 201, the node device inside the VNE receives a message carrying a destination IP address.

In this embodiment, the destination identification information is described by taking the destination IP address as an example, but it is not limited to the destination IP address. Wherein, the destination IP address has been described in detail in the first embodiment above, and will not be repeated here.

The node device in this embodiment may be an edge node device or an intermediate node device. If the node device is an edge node device, the edge node device receives the message sent by the network side; if the node device is an intermediate node device, the intermediate node device receives the message sent by other node devices inside the VNE. arts.

Step 202, the node device searches for an eFIB entry according to the destination IP address, and the eFIB entry includes the identifier of the system outbound interface corresponding to the destination IP address.

The node device finds the corresponding eFIB entry from the eFIB table stored by itself according to the destination IP address. Since the eFIB entry contains the system outgoing interface identifier, the node device can determine the VNE to the network side according to the system outgoing interface identifier. The system outbound interface to be used when sending the packet.

Step 203, the node device searches for an iFIB entry according to the system outbound interface identifier, and the iFIB entry includes the outbound interface identifier of the node.

The node device finds the corresponding iFIB entry from the iFIB table stored by itself according to the system outbound interface identifier contained in the eFIB table. Since the iFIB entry contains the outbound interface identifier of the node, the node device can The outbound interface identifier of the node determines the outbound interface of the node that needs to be used when the device on this node sends packets to other node devices inside the VNE.

Step 204, the node device judges whether the outgoing interface identifier of the local node is the same as the outgoing interface identifier of the system, if yes, execute step 205; if not, execute step 206.

Step 205, the node device sends a message to the network side through the outbound interface of the local node.

In this step, when the identification of the outgoing interface of the local node is the same as that of the outgoing interface of the system, the outgoing interface of the local node is the external connection port of the VNE (ie: the outgoing interface of the system). Send the message to the network side.

Step 206, the node device sends a message to the adjacent node device inside the VNE connected to the outbound interface of the local node.

In this step, when the identifier of the outgoing interface of the local node is different from that of the outgoing interface of the system, the outgoing interface of the local node is the internal connection interface of the VNE (that is, the non-system outgoing interface). At this time, the node equipment sends The message format can include the following two types:

In the first sending mode, the node device directly sends a message to an adjacent node through an outbound interface of the local node.

The second sending method.

The node device first encapsulates the internal routing label into the message, and the internal routing label contains the system outgoing interface identifier found by the node device. Then the node device sends the encapsulated message to the adjacent node through the outgoing interface of the local node.

In this embodiment, based on software technology, the destination identifier information and outgoing interface identifiers of multiple interconnected router devices are aggregated into FIB, so that the multiple interconnected router devices are aggregated into a virtual cluster routing system, and the packets are processed When forwarding, search for the eFIB entry containing the system outbound interface identifier according to the destination IP address in the message, then search for the iFIB entry containing the outbound interface identifier of the local node according to the system outbound interface identifier, and send the eFIB entry corresponding to the outbound interface identifier of the local node The adjacent node device connected to the outbound interface of the router sends the message, so that the entire virtual cluster routing system appears as a routing system externally, thereby simplifying the network topology and improving routing convergence and stability.

FIG. 3 is a schematic flow chart of a virtual cluster routing method provided by Embodiment 3 of the present invention. This embodiment is based on Embodiment 1 above, and describes a scenario where a node device only searches for iFIB entries. As shown in Figure 3, this embodiment may include the following steps:

It should be noted that this embodiment is the same as the above-mentioned embodiment 2, and the eFIB table and iFIB table are separated from each other, but the difference between this embodiment and the above-mentioned embodiment 2 is that the message received by the node device inside the VNE in this embodiment It is sent by other node devices inside the VNE, specifically, other node devices inside the VNE receive the message sent by the network side, follow the steps in the second embodiment above to find the system outbound interface ID, and will include the system outbound interface ID The internal routing label of the packet is encapsulated into the message, and the encapsulated message is sent to the node device in this embodiment according to the second sending method in step 206 in the above embodiment.

Before the steps of this embodiment are executed, the node devices inside the VNE need to generate the eFIB table and the iFIB table. The method for generating or updating the eFIB table and the iFIB table can refer to the description in the first embodiment above, and will not be repeated here.

Step 301, the node device inside the VNE receives the message encapsulated with the internal routing label.

In this embodiment, the destination identification information is described by taking an internal routing label as an example, but is not limited to the internal routing label. Wherein, the internal routing label has been described in detail in the first embodiment above, and will not be repeated here.

The node device in this embodiment may be an edge node device or an intermediate node device. Regardless of whether the node device is an edge node device or an intermediate node device, the messages received by the node device are all messages sent by other node devices inside the VNE.

Step 302, the node device parses the internal routing label encapsulated in the packet, and obtains the system outgoing interface identifier included in the internal routing label.

Steps 303 to 306 in this embodiment are the same as steps 203 to 206 in Embodiment 2 above, and will not be repeated here.

In this embodiment, according to the system outgoing interface identification encapsulated in the message, the iFIB entry containing the outgoing interface identification of the node is searched. Compared with the second embodiment above, the steps of searching for the eFIB entry are reduced, and the virtual cluster routing system is improved. s efficiency.

FIG. 4 is a schematic flowchart of a virtual cluster routing method provided in Embodiment 4 of the present invention. The virtual cluster routing method provided in this embodiment is described for a scenario of a load sharing mechanism in a VNE. As shown in Figure 4, this embodiment may include the following steps:

It should be noted that in this embodiment, there are two master node devices inside the VNE, and the two master node devices are mutually adjacent node devices. The master node devices have been described in the first embodiment above, and will not be repeated here. repeat. Wherein, a master node device is selected from the above two master node devices as a backup node device, and the backup node device is connected to the master node device through an internal connection port, and the backup node device is used to share part of the load of the master node device, and the The packets received by the VNE are sent to the network side.

Step 401, the master node device in the VNE sends a route sharing message to the backup node device.

In the process of Embodiment 1, the master node device can bear the current load, and the backup node device is in a waiting state. As the number of packets sent to the VNE in the external network continues to increase, the processing capability of the master node device can no longer meet the requirements of the network load. At this time, the master node device sends a route sharing message to the backup node device to instruct the backup node device to load sharing.

Step 402, the backup node device shares a part of the load of the master node device according to the load sharing mechanism, and sends the received part of the message to the network side.

Wherein, the load sharing mechanism is the basis for the standby node device to perform load sharing. The load sharing mechanism can determine the number of packets processed by the standby node device when performing load sharing. For example, the load sharing mechanism can determine the percentage of the number of packets processed by the standby node device to the total number of packets. When the load sharing mechanism is 30 %, when the load sharing process is executed, the standby node device will process 30% of the total number of messages.

Specifically, the process of performing load sharing by the standby node device is the same as the process in the first embodiment above, and will not be repeated here.

Further, when the number of packets received by the VNE decreases to the tolerance range of the primary node device, the backup node device stops routing packets and returns to the waiting state.

In this embodiment, when the number of packets received by the VNE is increasing, the load sharing process is executed to meet the requirements of the network load and improve the stability of the network.

FIG. 5 is a schematic flowchart of a virtual cluster routing method provided in Embodiment 5 of the present invention. The virtual cluster routing method provided in this embodiment is described for the active/standby redundancy scheme in the VNE. As shown in Figure 5, this embodiment may include the following steps:

It should be noted that the structure of the VNE in this embodiment is the same as the structure of the VNE in the fourth embodiment above, and will not be repeated here. The difference from the fourth embodiment above is that in this embodiment, the standby node device is used to replace the master node device to complete the virtual cluster routing method provided by the present invention when the master node device fails.

Step 501, the backup node device detects the working status of the master node device.

Wherein, during the process of executing the first embodiment above, the backup node device continuously detects the working status of the master node device.

Step 502, when the backup node device detects that the master node device fails, the backup node device replaces the master node device and sends the received message to the network side according to the master-standby switchover mechanism.

Wherein, the active-standby switchover mechanism is the basis for the standby node device to perform the active-standby switchover.

When the master node device is working normally, the standby node device is in the detection state. When the backup node device detects that the master node device fails, the backup node device replaces the master node device and sends the received message to the network side according to the active-standby switchover mechanism. Specifically, the backup node device replaces the master node device to send The process of sending the message on the network side is the same as the process in the first embodiment above, and will not be repeated here.

Further, when the backup node device replaces the master node device to send messages to the network side, the backup node device is still detecting the working status of the master node device, when the backup node device detects that the fault of the master node device has been eliminated and returns to the normal working state , perform active-standby switchover again, so that the standby node device returns to the detection state again, the master node device returns to the normal working state, and restarts the process in the first embodiment above.

In this embodiment, when the master node device fails, by executing the master-standby switchover process, the slave node device replaces the master node device to send the received message to the network side, thereby enhancing the reliability of the VNE.

FIG. 6 is a schematic structural diagram of a virtual cluster routing system provided in Embodiment 6 of the present invention. The virtual cluster routing system provided in this embodiment can be used to implement the flow of the virtual cluster routing method provided in the embodiment of the present invention shown in FIG. 1a to FIG. 5 . As shown in FIG. 6 , the virtual cluster routing system of this embodiment includes: a plurality of node devices 1 interconnected through ports, wherein the node device 1 includes: a receiving module 11 , a searching module 12 and a sending module 13 . Wherein, the receiving module 11 is used to receive the message carrying the purpose identification information; the search module 12 is used to search the FIB according to the purpose identification information to obtain the outbound interface identification of the node, and the FIB contains the purpose identification information and the outbound interface of the node. Correspondence between interface identifiers; the sending module 13 is configured to send the message to the adjacent node device connected to the outlet interface through the outgoing interface corresponding to the outgoing interface identifier of the local node.

Further, multiple node devices 1 in this embodiment are connected to other node devices through Ethernet, IP network or MPLS network.

When the virtual cluster routing system of this embodiment works, first, the receiving module 11 in the node device 1 receives a message sent by the network side, and the message carries destination identification information, and then, the search module 12 according to the destination identification information Find the FIB to obtain the outgoing interface identifier of the local node, wherein the FIB contains the correspondence between the destination identification information and the outgoing interface identifier of the local node, and the outgoing interface corresponding to the outgoing interface identifier of the local node is used to send the message, and the outgoing interface Adjacent node devices to which the node device 1 is connected. Finally, the sending module 13 sends the message to the adjacent node device through the outgoing interface.

In this embodiment, based on software technology, the destination identifier information and outgoing interface identifiers of multiple interconnected router devices are aggregated into FIB, so that the multiple interconnected router devices are aggregated into a virtual cluster routing system, and the packets are processed When forwarding, the search module 12 searches for the FIB according to the destination identification information carried in the message, obtains the outbound interface identifier of the node, and sends the message through the outbound interface corresponding to the outbound interface identifier of the node by the sending module 13, so that the entire virtual cluster routing system Externally, it appears as a routing system, which simplifies the network topology, simplifies operation and maintenance management, and improves routing convergence and stability.

Further, the node device 1 of the virtual cluster routing system in this embodiment further includes: a FIB generating module 21 for merging eFIB and iFIB to generate FIB.

If the node device 1 inside the virtual cluster system of this embodiment generates or updates the eFIB in a distributed and dynamic generation manner, the FIB generating module 21 further includes: a protocol receiving sub-module 14 and an eFIB generating sub-module 15 . Among them, the protocol receiving submodule 14 is used to receive the neighbor information sent by the network side, and the neighbor information includes the topology information of the network side; the eFIB generating submodule 15 is used to generate or update the eFIB according to the topology information in the neighbor information , the eFIB contains the correspondence between the destination IP address and the system outbound interface identifier.

At this time, the FIB generation module 21 may further include: a protocol synchronization sub-module 16, configured to synchronize neighbor information with other node devices inside the virtual cluster routing system.

The node device 1 inside the virtual cluster routing system in this embodiment can also generate or update the eFIB in a centralized and dynamic generation mode. At this time, the protocol synchronization submodule 16 of the FIB generation module 21 can also be used to communicate with the virtual cluster routing Other node devices in the system synchronize eFIB.

The node device 1 inside the virtual cluster routing system in this embodiment can also generate or update the eFIB by means of external configuration generation. At this time, the FIB generation module 21 can also include: eFIB receiving sub-module 17 for receiving network management equipment settings or the eFIB sent by other node devices inside the virtual cluster routing system, the eFIB includes the correspondence between the destination IP address and the system outgoing interface identifier.

Furthermore, if the node device 1 inside the virtual cluster routing system in this embodiment generates an iFIB in a static configuration mode, the FIB generating module 21 also includes: the iFIB receiving sub-module 18 is used to receive the set or The iFIB sent by other nodes inside the virtual cluster routing system.

If the node devices 1 inside the virtual cluster routing system in this embodiment generate iFIBs in a distributed and dynamic manner, the FIB generation module 21 further includes: an iFIB detection submodule 19 and an iFIB generation submodule 20 . Wherein, the iFIB detection sub-module 19 is used to dynamically detect the topology inside the virtual cluster routing system; the iFIB generation sub-module 20 is used to generate or update the iFIB of the node according to the topology.

FIG. 7 is a schematic structural diagram of the virtual cluster routing system provided by Embodiment 7 of the present invention. This embodiment is based on the virtual cluster routing system provided by Embodiment 6 above, and can be used to realize the virtual cluster provided by the embodiment of the present invention shown in FIG. 2 The flow of the routing method. As shown in FIG. 7 , the search module 12 in the node device 1 inside the virtual cluster routing system of this embodiment further includes: a first search unit 121 and a second search unit 122 . Wherein, the first search unit 121 is configured to search the eFIB according to the destination IP address when the destination identification information is the destination IP address, and obtain the system outgoing interface identifier, and the eFIB includes the correspondence between the destination IP address and the system outgoing interface identifier. Relationship; the second search unit 122 is configured to search the iFIB according to the system outbound interface identifier to obtain the outbound interface identifier of the node, and the iFIB includes the corresponding relationship between the outbound interface identifier of the system and the outbound interface identifier of the node.

Further, the sending module 13 of the node device 1 in this embodiment is specifically configured to send a message to the network side through the outbound interface of the local node;

Furthermore, the sending module 13 of the node device 1 in this embodiment may include: an encapsulation unit 131 and a sending unit 132 . Wherein, the encapsulating unit 131 is used for encapsulating the internal routing label into the message, and the internal routing label contains the system outgoing interface identifier; The node device sends the encapsulated message.

When the virtual cluster routing system of this embodiment works, first, after the receiving module 11 receives the message carrying the destination IP address, the first search unit 121 in the search module 12 searches the eFIB according to the destination IP address to obtain the system outbound interface Identification, wherein, the eFIB contains the correspondence between the destination IP address and the system outgoing interface identification. Afterwards, the second searching unit 122 searches the iFIB according to the system outbound interface identifier to obtain the outbound interface identifier of the local node, wherein the iFIB includes the outgoing interface identifier of the local node. Finally, when the identifier of the outgoing interface of the local node is the same as that of the outgoing interface of the system, the sending module 13 sends the message to the network side through the outgoing interface of the local node.

Further, when the identifier of the outgoing interface of the local node is different from the identifier of the outgoing interface of the system, the encapsulation unit 131 in the sending module 13 may encapsulate the internal routing label into the message, and the internal routing label contains the system outgoing interface The interface identifier, and then the sending unit 132 sends the message encapsulated by the encapsulating unit 131 to the adjacent node device connected to the outgoing interface of the current node.

In this embodiment, based on software technology, the destination identifier information and outgoing interface identifiers of multiple interconnected router devices are aggregated into FIB, so that the multiple interconnected router devices are aggregated into a virtual cluster routing system, and the packets are processed When forwarding, the first search unit 121 in the node device 1 searches for the eFIB containing the system outbound interface identifier according to the destination IP address in the message, and then the second search unit 122 searches for the eFIB containing the outbound interface identifier of the current node according to the system outbound interface identifier. The iFIB of the interface identification, and the sending module 13 sends the message to the adjacent node device connected to the outgoing interface of the local node, so that the entire virtual cluster routing system appears as a routing system externally, thereby simplifying the network topology and improving routing convergence sex and stability.

FIG. 8 is a schematic structural diagram of the virtual cluster routing system provided by the eighth embodiment of the present invention. Based on the virtual cluster routing system provided by the sixth embodiment of the present invention, it can be used to realize the virtual cluster routing provided by the embodiment of the present invention shown in FIG. 3 The flow of the method. As shown in FIG. 8 , the difference between the node device 1 inside the virtual cluster routing system of this embodiment and the node device 1 in the seventh embodiment above is that the search module 12 in this embodiment includes: an analysis unit 123 and a third search unit 124. Wherein, the parsing unit 123 is used to parse the internal routing label when the destination identification information is an internal routing label, and obtain the system outgoing interface identifier contained in the internal routing label; the third searching unit 124 is used to The identifier looks up the iFIB, and the iFIB includes the identifier of the outbound interface of the local node of the message.

When the virtual cluster routing system of this embodiment is working, first, after the receiving module 11 receives the message encapsulated with the internal routing label, the parsing unit 123 in the search module 12 analyzes the internal routing label to obtain the information contained in the internal routing label. System outgoing interface identification, after that, the third search unit 124 searches for iFIB according to the system outgoing interface identification, which contains the outgoing interface identification of this node in the iFIB, and finally, the sending module 13 can send a message according to the outgoing interface identification of this node Refer to the description in the seventh embodiment above, and details are not repeated here.

In this embodiment, the parsing unit 123 parses the system outbound interface identifier from the message, and the third search unit 124 finds out the iFIB containing the outbound interface identifier of the node according to the system outbound interface identifier, compared with the seventh embodiment above In other words, the steps of searching eFIB are reduced, and the efficiency of the virtual cluster routing system is improved.

FIG. 9 is a schematic structural diagram of a virtual cluster routing system provided in Embodiment 9 of the present invention. The virtual cluster routing system provided in this embodiment can be used to implement the flow of the virtual cluster routing method provided in the embodiment of the present invention shown in FIG. 4 . As shown in Figure 9, the node device 1 in the virtual cluster routing system in this implementation is specifically used to share the partial load sent by the adjacent node device 2 inside the virtual cluster routing system according to the load sharing mechanism, and send the received message to sent by the network side.

When the virtual cluster routing system of this embodiment is working, when the adjacent node 2 inside the virtual cluster routing system is working normally, the node device 1 is in a waiting state. As the number of messages sent to the adjacent node device 2 in the external network continues to increase, the adjacent node device 2 can no longer meet the requirements of the network load. At this time, the adjacent node device 2 sends a route sharing message to the node device 1, indicating that the Node device 1 performs load sharing. After node device 1 receives the route sharing message, it shares part of the load of adjacent node device 2 according to the load sharing mechanism, and sends the received message to the network side.

In this embodiment, the node device 1 shares part of the load of the adjacent node device 2, which meets the requirement of the network load and improves the stability of the network.

The virtual cluster routing system provided by Embodiment 10 of the present invention has the same structure as the virtual cluster routing system provided by Embodiment 9 above. The specific structure can be referred to in FIG. 9 . The virtual cluster routing system provided in this embodiment can be used to realize the The flow of the virtual cluster routing method provided by the embodiment of the present invention is shown. As shown in Figure 9, the node device 1 in the virtual cluster routing system of this embodiment is specifically used to replace the adjacent node device 2 inside the virtual cluster routing system to receive messages according to the active-standby switchover mechanism when the adjacent node device 2 fails , and send it to the network side.

When the virtual cluster routing system of this embodiment is working, the node device 1 continuously detects the working status of the adjacent node device 2 inside the virtual cluster routing system. When the adjacent node device 2 is working normally, the node device 1 is in the detection state. When the adjacent node device 2 fails, the node device 1 replaces the adjacent node device 2 and sends the received message to the network side according to the active/standby switchover mechanism.

In this embodiment, when the adjacent node device 2 fails, the node device 1 replaces the adjacent node device 2 and sends the received message to the network side by performing the active/standby switchover process, thereby enhancing the reliability of the virtual cluster routing system.

Those of ordinary skill in the art can understand that all or part of the steps for realizing the above-mentioned method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the It includes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other various media that can store program codes.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (25)

1. A virtual cluster routing method, characterized in that it is suitable for a virtual cluster routing system, and the virtual cluster routing system includes a plurality of node devices interconnected by ports; comprising: The node device receives a message carrying destination identification information; Searching the forwarding information base FIB according to the target identification information to obtain the outgoing interface identification of this node, the FIB includes the corresponding relationship between the purpose identification information and the outgoing interface identification of this node; Sending the message to the adjacent node device connected to the outgoing interface through the outgoing interface corresponding to the outgoing interface identifier of the local node. 2. The method according to claim 1, wherein when the target identification information is a target IP address, searching for the FIB according to the target identification information comprises: The node device searches the external forwarding information base eFIB according to the destination IP address, and obtains the system outgoing interface identifier, and the eFIB includes the corresponding relationship between the destination IP address and the system outgoing interface identifier; The node device searches the internal forwarding information base iFIB according to the system outgoing interface identifier to obtain the local node outgoing interface identifier, and the iFIB contains the corresponding relationship between the system outgoing interface identifier and the local node outgoing interface identifier. 3. The method according to claim 1, wherein when the destination identification information is an internal routing label, searching for the FIB according to the destination identification information comprises: The node device parses the internal routing label, and obtains the system outgoing interface identifier contained in the internal routing label; The node device searches for an iFIB according to the system outgoing interface identifier, and the iFIB includes the local node outgoing interface identifier of the message. 4. The method according to any one of claims 1-3, wherein when the outgoing interface of the local node is a system outgoing interface, the message is sent through the outgoing interface corresponding to the outgoing interface identifier of the local node The adjacent node devices connected to the outgoing interface include: Send the message to an adjacent node device inside the virtual cluster routing system connected to the outbound interface through the outbound interface corresponding to the outbound interface identifier of the local node 5. The method according to any one of claims 1-3, wherein when the outgoing interface of the local node is a non-system outgoing interface, the report is sent through the outgoing interface corresponding to the outgoing interface identifier of the local node. The adjacent node devices connected to the outgoing interface include: Send the message to the adjacent node device on the network side connected to the outgoing interface through the outgoing interface corresponding to the outgoing interface identifier of the local node 6. The method according to any one of claims 1-3, wherein when the outbound interface corresponding to the outbound interface identifier of the local node is a non-system outbound interface, the pass corresponds to the outbound interface identifier of the local node The outgoing interface sending the message to the adjacent node device connected to the outgoing interface includes: The node device encapsulates an internal routing label into the message, and the internal routing label includes the system outgoing interface identifier; The node device sends an encapsulated message to an adjacent node device inside the virtual cluster routing system connected to the outbound interface through the outbound interface corresponding to the outbound interface identifier of the local node 7. The method according to claim 2, further comprising: The node device receives neighbor information sent by the network side, where the neighbor information includes topology information on the network side; The node device generates or updates an eFIB according to the topology information in the neighbor information, and the eFIB includes the correspondence between the destination IP address and the system outgoing interface identifier. 8. The method according to claim 7, further comprising: The node device synchronizes neighbor information with other node devices inside the virtual cluster routing system; or The node device synchronizes the eFIB with other node devices inside the virtual cluster routing system. 9. The method of claim 2, further comprising: The node device receives the eFIB set by the network management device or sent by other node devices inside the virtual cluster routing system, and the eFIB includes the correspondence between the destination IP address and the system outgoing interface identifier. 10. The method of claim 2, further comprising: The node device receives the iFIB set by the network management device or sent by other nodes inside the virtual cluster routing system. 11. The method of claim 2, further comprising: The node device dynamically detects the internal topology of the virtual cluster routing system; The node device generates or updates the iFIB of the node according to the topology. 12. The method of claim 1, further comprising: According to the load sharing mechanism, the node device shares part of the load sent by the adjacent node devices inside the virtual cluster routing system, and sends the received message to the network side. 13. The method according to claim 1, wherein, when an adjacent node device within the virtual cluster routing system fails, further comprising: According to the active-standby switchover mechanism, the node device replaces the adjacent node device inside the virtual cluster routing system to receive the message and send it to the network side. 14. A virtual cluster routing system, characterized in that, the virtual cluster routing system includes: a plurality of node devices interconnected through ports, and the node devices also include: A receiving module, configured to receive a message carrying purpose identification information; A search module, configured to search the forwarding information base FIB according to the target identification information, and obtain the outgoing interface identification of the current node, and the FIB includes the corresponding relationship between the target identification information and the outgoing interface identification of the current node; A sending module, configured to send the message to an adjacent node device connected to the outgoing interface through the outgoing interface corresponding to the outgoing interface identifier of the local node. 15. The virtual cluster routing system according to claim 14, wherein the search module specifically comprises: The first search unit is used to search the external forwarding information base eFIB according to the destination IP address when the destination identification information is the destination IP address, and obtain the system outgoing interface identifier, and the eFIB includes the destination IP address and the system outgoing interface identifier corresponding relationship; The second search unit is configured to search the internal forwarding information base iFIB according to the system outgoing interface identifier to obtain the local node outgoing interface identifier, and the iFIB includes the corresponding relationship between the system outgoing interface identifier and the local node outgoing interface identifier. 16. The virtual cluster routing system according to claim 14, wherein the search module specifically includes: A parsing unit, configured to parse the internal routing label when the destination identification information is an internal routing label, and obtain the system outgoing interface identification contained in the internal routing label; A searching unit, configured to search for an iFIB according to the system outgoing interface identifier, where the iFIB contains the local node outgoing interface identifier of the message. 17. The virtual cluster routing system according to claim 14, wherein the sending module specifically comprises: An encapsulation unit, configured to encapsulate an internal routing label into the message, where the internal routing label contains the system outgoing interface identifier; The sending unit is configured to send the encapsulated message to the adjacent node device connected to the outgoing interface of the local node. 18. The virtual cluster routing system according to claim 14, wherein the node device further comprises: The FIB generating module is used for merging eFIB and iFIB to generate FIB. 19. The virtual cluster routing system according to claim 18, wherein the FIB generating module comprises: A protocol receiving submodule, configured to receive neighbor information sent by the network side, where the neighbor information includes topology information of the network side; The eFIB generating submodule is configured to generate or update an eFIB according to the topology information in the neighbor information, and the eFIB includes the correspondence between the destination IP address and the system outgoing interface identifier. 20. The virtual cluster routing system according to claim 19, wherein the FIB generation module further comprises: A protocol synchronization submodule, configured to synchronize neighbor information with other node devices inside the virtual cluster routing system; or The protocol synchronization sub-module is specifically used to synchronize eFIB with other node devices inside the virtual cluster routing system. 21. The virtual cluster routing system according to claim 18, wherein the FIB generating module comprises: The eFIB receiving sub-module is used to receive the eFIB set by the network management device or sent by other node devices inside the virtual cluster routing system, the eFIB includes the corresponding relationship between the destination IP address and the system outgoing interface identifier. 22. The virtual cluster routing system according to claim 18, wherein the FIB generating module comprises: The iFIB receiving sub-module is used to receive the iFIB set by the network management device or sent by other nodes inside the virtual cluster routing system. 23. The virtual cluster routing system according to claim 18, wherein the FIB generation module comprises: The iFIB detection submodule is used to dynamically detect the topology inside the virtual cluster routing system; the iFIB generation submodule is used to generate or update the iFIB of the node according to the topology. 24. The virtual cluster routing system according to claim 14, wherein the node device is specifically configured to share the partial load sent by the adjacent node devices inside the virtual cluster routing system according to the load sharing mechanism, and will receive The message is sent to the network side. 25. The virtual cluster routing system according to claim 14, wherein the node device is specifically configured to replace the virtual Adjacent node devices inside the cluster routing system receive the message and send it to the network side.

Images