KR20040042263A - Mesh Group Establish Method In Communication Network - Google Patents

Abstract

PURPOSE: A mesh group setting method in a communication network is provided to automatically set a mesh group by enabling each node having point-to-point connection to search full connectivity with an adjacent node. CONSTITUTION: Each node searches interface forms of neighbor nodes recorded in an interface table(S21) to check whether the nodes have point-to-point connection(S22). If a node has point-to-point connection with an adjacent node, the node searches an interface state with the corresponding neighbor node(S23) to check whether point-to-point connection with the neighbor node has been enabled(S24). If the point-to-point connection has been enabled, the node exchanges a hello packet with the corresponding neighbor node to make adjacency relation and checks an ID of the neighbor node, namely, an adjacent node(S25). The node exchanges a link state PDU with the adjacent node which has been made the adjacency relation(S26). The two mutually adjacent nodes check whether they have the same adjacent nodes(S27). If it is checked that they have the same adjacent nodes, the node registers the corresponding adjacent node with which it has exchanged the link state PDU as a mesh group(S28).

Description

Mesh Group Establish Method In Communication Network

The present invention relates to a mesh group setting in a communication network. In particular, a mesh group setting in a communication network in which each node having a point-to-point connection is searched for full connectivity with an adjacent node and automatically establishes a mesh group It is about a method.

In general, in a point-to-point network such as Asynchronous Transfer Mode (ATM) or Frame Relay, which includes a plurality of nodes coupled together, redundant packets between nodes in a routing protocol. In order to reduce the transmission, we define an IS-IS mesh group (hereinafter referred to as a mesh group), which is the point of each node with full connectivity in the network. It is defined as a set of dot circuits. In other words, it refers to a set of point-to-point circuits of each node in a point-to-multipoint fully connected network.

For example, when four nodes have a full mesh topology configuration, that is, when four nodes form a fully connected network, as shown in FIG. 1, the node 1 is in a link state PDU. When a link state protocol data unit is transmitted, Node 1 receiving the corresponding link state PDU floods the link state PDU to all lines except the source circuit, that is, the I12, I13, and I14 lines. Done.

Then, the node 2 receiving the link state PDU through the I12 line will flood the corresponding link state PDUs to the node 3 and the node 4 through the I23 and I24 lines except the I12 line, where the node 3 and the node 4 are already node 1 The link state PDU has been received from.

Therefore, conventionally, when N nodes are completely meshed, N-2 additional transmissions for link state PDUs occur, which causes a problem of wasting bandwidth and system processing.

Therefore, in order to compensate for the shortcomings of the IS-IS protocol according to the above-described full access network configuration, a mesh group has been defined in the related art, and each mesh group has full connectivity to each node composed of ATM or frame relay point-to-point connection. In this case, the network administrator directly configures the packet for efficient packet transmission.

The network group configured by the network administrator is defined as a set of point-to-point circuits of each node having full connectivity, where each circuit is a mesh group enabled indicating a mesh setting state and an identifier of the mesh group. It has two attributes called meshGroup.

In addition, the mesh group enable attribute indicates the states of mesh set, mesh inactive, and mesh stopped according to the state of the mesh group, and the mesh group attribute is set by the mesh group attribute. The state indicates an identifier of a mesh group that is meaningful only in a state. Here, circuits having the same mesh group identifier indicating a mesh setting state may be referred to as one mesh group.

Therefore, each node having a point-to-point connection between each other and configuring a fully connected network does not flood the corresponding link state PDUs to the circuits in the mesh down state when the node receives the link state PDUs. Since the link state PDU is not flooded to the line to which the same mesh group belongs to the link state PDU, the flooding of the link state PDU can be reduced by the above-described mesh group setting.

However, in the point-to-point environment where the mesh group needs to be set as described above, the mesh group is directly set by the network administrator, so the state and configuration of the network must be relied on only by the administrator's monitoring and understanding. In addition, there is no problem in setting up a mesh group by a network manager in a relatively simple fully connected network environment. However, when a network is gradually expanded and the management elements (nodes, links, etc.) of the manager increase and become complicated, it is difficult to set up an efficient mesh group. there was.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and its object is to provide full connectivity through exchange of hello packet and link state PDUs between nodes having a point-to-point connection in a fully connected network in which the IS-IS protocol is operated. This is to search for and to automatically register neighboring nodes with full connectivity as a mesh group.

Another object of the present invention is to register the mesh group by searching for complete connectivity between each node, so that the mesh group can be automatically set up without a topology analysis by the network manager, and an efficient mesh group setting is possible even when the network is expanded. To make it possible.

1 is a diagram illustrating a network configuration between nodes with general full connectivity.

FIG. 2 is a flowchart illustrating an operation of setting a mesh group through full connectivity search between nodes in a fully connected network environment according to the present invention. FIG.

A feature of the present invention for achieving the object as described above, the method comprising the steps of checking the neighbor node ID after exchanging the hello packet between each node with the point-to-point connection enabled to form a neighbor relationship with the neighbor node; Exchanging a link state PDU with the neighbor node to determine whether the neighbor nodes include the same neighbor nodes; The present invention provides a method for establishing a mesh group in a communication network including registering a neighbor node that exchanges the link state PDUs and setting a mesh group when the same neighboring nodes are included.

The checking of the neighbor node ID may include: checking whether an interface table has a point-to-point connection by searching an interface table; If the point-to-point connection has a point-to-point connection with a corresponding neighbor node is enabled; If the point-to-point connection with the neighbor node is enabled, the step of checking the ID of the neighbor node of the neighbor node after establishing a neighbor relationship by exchanging hello packets.

In addition, the process of exchanging the link state PDUs to determine whether they include the same neighbor nodes may include determining whether the IS neighbor node information contained in the variable length field of the link state PDU received from the neighbor node is consistent with its IS neighbor node information. It is characterized by checking whether or not including the same adjacent nodes to each other.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In a fully connected network operating the IS-IS protocol according to the present invention, each node, not the network manager, automatically sets a mesh group that is set to reduce flooding of link state PDUs between nodes having a point-to-point connection. In this case, the management elements for automatically searching for complete connectivity in each node include interface type (point-to-point or broadcast) and interface status (enable / disable). Neighbor node ID is required, and various management conditions are searched with these management elements to determine complete connectivity.

For reference, a fully-connected point-to-point network is composed mainly of ATM or frame relay and is similar to Non-Broadcast Multi Access of Open Shortest Path First (OSPF) protocol. It's a network.

For example, in a fully connected network environment in which four nodes have a full mesh topology configuration, as shown in FIG. 1, each node includes a neighbor node and a hello packet near each other. You will have an adjacency node relationship that communicates with and receives the connection.

After establishing an adjacent relationship with a node directly connected to each other, each node exchanges link state information by exchanging link state PDUs thereof, so that each node has a link state DB for each node and has a routing information base. Through this procedure, each node can know its network configuration by searching its neighbor node information and the link state DB received from the neighbor node.

That is, an operation of setting a mesh group through full connectivity search between nodes in a fully connected network environment according to the present invention will be described with reference to FIG. 2.

First, each node searches for an interface type with a neighbor node recorded in an interface table storing interface information about neighbor nodes (step S21), and checks whether or not there is a point-to-point connection (step S22).

At this time, as a result of searching the interface type with the neighboring node, if there is a point-to-point connection, by checking the interface state with the corresponding neighboring node (step S23), it is checked whether the point-to-point connection with the neighboring node is enabled. (Step S24).

As a result of checking the interface state with the neighbor node, when the point-to-point connection is enabled, a neighbor relationship is established by exchanging hello packets with the corresponding neighbor node. In this case, the ID of the neighbor node, which is the neighbor node having the neighbor relationship, is established. Is checked (step S25).

Thereafter, each node exchanges a link state PDU with an adjacent node which has an adjacent relationship with itself. In this case, identification information about adjacent nodes having an adjacent relationship with itself is displayed in a variable length field of the link state PDU. The IS neighbor node ID is recorded and exchanged (step S26).

Therefore, each node receiving the link state PDU from the neighbor node checks the IS neighbor node IDs recorded in the variable length field of the corresponding link state PDU and includes the same neighbor nodes, that is, the link state PDU received from the neighbor node. It is checked whether the IS neighbor node information contained in the variable length field matches the IS neighbor node information of their own to check whether they include the same neighbor nodes (step S27). If so, the neighbor node that exchanged the current link state PDU is registered as a mesh group (step S28).

In other words, each node automatically checks whether a neighboring node that is adjacent to each other as an enable point-to-point connection and each includes the same neighboring nodes, thereby automatically generating a mesh group including adjacent nodes having full connectivity. It can be set.

On the other hand, as a result of searching for the interface type with the neighbor node in step S22 as a result of having a broadcast connection rather than a point-to-point connection, or as a result of searching for the interface state with the neighbor node in step S24, the point-to-point connection is disabled. Or, if it is determined in step S27 that the IS neighbor node IDs recorded in the link state PDU received from the neighbor node do not include the same neighbor nodes, the corresponding neighbor node (or neighbor node) is a mesh group. Since it is not a target, it ends without performing the above-described mesh group registration.

Herein, a network environment as shown in FIG. 1 is described with reference to the accompanying drawings, in which each node exchanges link state PDUs with neighbor nodes to determine whether they include the same neighbor nodes.

First, assuming that nodes 1 to 4 are neighbor nodes having an adjacent relationship as point-to-point connections in an enabled state, each node exchanges link state PDUs with neighbor nodes, where node 1 is in link state. Node 2, 3, and 4 information is sent to the PDU as its IS neighbor node ID, and node 2 carries information of nodes 1, 3, and 4 as its IS neighbor node to the link state PDU.

In addition, node 1 checks IS neighbor node information included in a link state PDU received from node 2, which is an adjacent node, and node 1, which is an adjacent node, to determine whether node 1 includes identical neighbor nodes. Node 2 and Node 2 check the IS neighbor node information on the link state PDU. As a result, nodes 3 and 4 are identically included as nodes adjacent to each other. Therefore, neighbor nodes that exchange the current link state PDU are registered as mesh groups. You can also set up a mesh group automatically by performing this procedure with other neighboring nodes.

In addition, the embodiments according to the present invention are not limited to the above-described embodiments, and various alternatives, modifications, and changes can be made within the scope apparent to those skilled in the art.

As described above, the present invention searches for complete connectivity through the exchange of hello packet and link state PDU between nodes having a point-to-point connection in a fully connected network in which the IS-IS protocol is operated. By registering neighboring nodes as a mesh group, a mesh group can be automatically set up without a topology analysis by a network manager. Accordingly, an efficient mesh group can be set up even if the network is expanded.

Claims (3)

Exchanging hello packets between each node for which point-to-point connection is enabled to establish a neighbor relationship with a neighbor node, and then checking neighbor node IDs; Exchanging a link state PDU with the neighbor node to determine whether the neighbor nodes include the same neighbor nodes; And setting a mesh group by registering neighbor nodes that exchange the link state PDUs if they include the same neighbor nodes. The method of claim 1, The checking of the neighbor node ID may include: searching an interface table to determine whether a neighbor node has a point-to-point connection; If the point-to-point connection has a point-to-point connection with a corresponding neighbor node is enabled; When the point-to-point connection with the neighbor node is enabled, after establishing a neighbor relationship by exchanging hello packets, checking the ID of the neighbor node as the neighbor node; . The method of claim 1, The process of exchanging the link state PDUs to check whether they include the same neighbor nodes is performed by checking whether IS neighbor node information contained in the variable length field of the link state PDU received from the neighbor node matches its IS neighbor node information. Method of setting a mesh group in a communication network, characterized in that it comprises the same adjacent nodes to each other.