KR102657778B1 - Apparatus for managing network using clustering by failure event, and method thereof - Google Patents

Abstract

The present invention relates to a network management device and method. Instead of managing network failures on a node/link basis in the past, the entire network is separated into clusters to determine and manage failures in the entire network due to each cluster failure, making it faster and more efficient. It has the advantage of being able to efficiently manage large-scale networks.

Description

Network management device and method using clustering by failure event {APPARATUS FOR MANAGING NETWORK USING CLUSTERING BY FAILURE EVENT, AND METHOD THEREOF}

The present invention relates to network management, and particularly to network management when a failure event occurs.

Many systems in modern society have a network structure through a network of connections. For example, communication networks, water networks, power networks, and social networks all have a network structure. This network system is made up of numerous entities (nodes), and these entities are increasing in modern times.

Conventionally, networks have been managed by managing each entity (node) that constitutes the network. However, in an era where infrastructure systems are expanding and infinite data is pouring out, there is a problem that it is no longer difficult to determine whether there is a system failure by analyzing the status of each node in the network or analysis of neighboring nodes.

The inventors of the present invention have made research efforts to solve the difficulties of the network node management problem of the prior art. After much effort, the present invention was completed to complete a device and method that can quickly and accurately detect the impact of a failure event on the entire network system unit by dividing the network into cluster units and analyzing the network from a clustering perspective.

The purpose of the present invention is to quickly and accurately detect the impact of a failure event on the entire system by understanding the network structure in clustering units.

Meanwhile, other unspecified purposes of the present invention will be additionally considered within the scope that can be easily inferred from the detailed description and effects below.

The network management device according to the present invention,

a network structure management unit that determines the network structure based on node information and link information connecting nodes; Node/link event management unit that collects failure events of nodes or links; a network clustering unit that reconfigures the network structure in clusters when a failure of the node or link occurs; and a network failure determination unit that determines whether the entire network has a failure in the reconfigured cluster unit.

The network failure determination unit is characterized in that it determines whether the unit cluster has a failure based on whether the reconfigured unit cluster includes a seed node.

The network failure determination unit is characterized in that it creates a failure cluster slip including the number of nodes included in the failure cluster among the reconfigured clusters and a node list.

The network failure determination unit is characterized in that when the failure node or link collected by the node/link event management unit is included in the failure cluster slip, the failure node or link event is merged into the failure cluster slip.

A network management method according to another embodiment of the present invention:

collecting failure events of nodes or links; Reorganizing the network into clusters separated by failure of the node or link; and determining whether the entire network has a failure in the reconfigured cluster unit.

The step of determining whether the entire network has a failure is characterized by determining whether the unit cluster has a failure based on whether the reconfigured unit cluster includes a seed node.

The step of determining whether the entire network has a failure is characterized by creating a failure cluster slip including the number of nodes included in the failed cluster among the reconfigured clusters and a node list.

After collecting the failure event of the node or link, determining whether the node or link where the failure event occurred is included in the failure cluster slip, further comprising: the node or link where the failure event occurred is included in the fault cluster slip If included in , the failed node or link event is merged into the fault cluster slip.

According to the present invention, it is possible to quickly identify failures in the entire system rather than node-level failures, which has the effect of resolving failures in the entire network system more quickly.

Meanwhile, it is to be added that even if the effects are not explicitly mentioned herein, the effects described in the following specification and their potential effects expected from the technical features of the present invention are treated as if described in the specification of the present invention.

1 is a structural diagram of a network management device according to a preferred embodiment of the present invention.
2 to 4 are examples of networks managed by a network management device according to a preferred embodiment of the present invention.
Figure 5 is a flowchart of a network management method according to another preferred embodiment of the present invention.
※ The attached drawings are intended as reference for understanding the technical idea of the present invention and are not intended to limit the scope of the present invention.

Hereinafter, with reference to the drawings, we will look at the configuration of the present invention guided by various embodiments of the present invention and the effects resulting from the configuration. In describing the present invention, if it is determined that related known functions may unnecessarily obscure the gist of the present invention as they are obvious to those skilled in the art, the detailed description thereof will be omitted.

Terms such as 'first' and 'second' may be used to describe various components, but the components should not be limited by the above terms. The above terms may be used only for the purpose of distinguishing one component from another. For example, the 'first component' may be named 'the second component' without departing from the scope of the present invention, and similarly, the 'second component' may also be named 'the first component'. You can. Additionally, singular expressions include plural expressions, unless the context clearly dictates otherwise. Unless otherwise defined, terms used in the embodiments of the present invention may be interpreted as meanings commonly known to those skilled in the art.

Hereinafter, with reference to the drawings, we will look at the configuration of the present invention guided by various embodiments of the present invention and the effects resulting from the configuration.

1 is a structural diagram of a network management device according to a preferred embodiment of the present invention.

The network management device 10 according to the present invention includes a network structure management unit 20, a node/link event management unit 30, a network clustering unit 40, and a network failure determination unit 50.

The network management device 10 according to the present invention includes one or more processors and memory. The memory may include program codes and data for driving a processor to implement the present invention.

Figure 2 shows an example of a network managed by the network management device 10 according to the present invention.

Figure 2(a) shows an example of a network consisting of 6 nodes. The network consists of 6 nodes and 6 links connecting nodes, with node 1 serving as a seed node. In other words, even if there is a failure of some nodes or links, if the nodes can be connected to Node 1, the seed node, it can be judged that there is no effect on the entire network.

For example, if node 4 fails in (b) of Figure 2, node 3 can still connect to node 1, the seed node, but node 2, node 5, and node 6 can connect to node 1, the seed node. Therefore, a failure occurs in the entire network.

On the other hand, in the case shown in (c) of FIG. 2, even if a failure occurs in node 3 or the links connected to node 3, node 2, node 5, and node 6 can still be connected to node 1, which is the seed node, through node 4. In other words, the failure of Node 3 can be seen as having no effect on the entire network except for Node 3 where the failure occurred.

Therefore, the present invention is intended to determine the impact of individual node/link failure events on the entire network by reorganizing the network on a cluster basis.

The network structure management unit 20 may include a node information DB 22 and a link information DB 24. The network structure management unit 20 identifies and manages the network structure based on node/link information.

For a network such as (a) in FIG. 2, the node information DB 22 may include node information as shown in Table 1 below.

node_no node_description link_degree cluster_no seed_yn One A 2 One One 2 B One One 0 3 C 2 One 0 4 D 3 One 0 5 E 2 One 0 6 F 2 One 0

The node information DB 22 may include a node number (node_no), node description (node_description), link number (link_degree), cluster number (cluster_no), and whether or not it is a seed node (seed_yn).

The node number represents the serial number of the node.

The node description stores the node's unique features and necessary explanations.

The number of links indicates the number of links connected to each node.

In the case of node 2, only node 6 is connected, so the number of links is 1, and node 4 is connected to node 1, node 3, and node 5, so the number of links is 3.

The present invention understands the network in cluster units. Therefore, nodes that are directly or indirectly connected to each other are identified as one cluster.

For example, in (a) of Figure 2, all nodes are connected to each other, so it consists of one cluster, and in the case of (b) of Figure 2, node 1 and node 3 are connected to each other, node 2, and node 5. and Node 6 are connected to each other but not to Node 1 or Node 3, so Node 1 and Node 3 form one cluster, and Node 2, Node 5, and Node 6 form another cluster. In other words, the entire network is reorganized into two clusters due to the failure of node 4.

Additionally, the link information DB 24 may include link information as shown in Table 2 below.

node_no connected_node One [3,4] 2 [6] 3 [1,4] 4 [1,3,5] 5 [4,6] 6 [2,5]

The link information DB 24 includes a node number (node_no) and a list of all nodes to which the node is connected (connected_node).

In the example of the network shown in (a) of Figure 2, node 1 is connected to node 3 and node 4, node 2 is connected only to node 6, and node 4 is connected to node 1, node 3, and node 5.

The network structure management unit 20 can manage the structure of the entire network by managing the node information DB 22 and the link information DB 24 in this way.

The node/link event management unit 30 collects information on nodes or links in which a failure has occurred. This is to determine which node or link in the entire network has failed and its impact on the entire network.

When events are collected by the node/link event management unit 30, the network clustering unit 40 reorganizes the network in cluster units.

The network clustering unit 40 forms a cluster using the node information DB 22 and link information DB 24 of the network structure management unit 20 and stores the information in the cluster information DB 42.

For this purpose, the network clustering unit 40 may include a cluster information DB 42. Table 3 below shows an example of information stored in the cluster information DB 42.

cluster_no seed_yn cluster_size One One 6

Since the entire network as shown in (a) of Figure 2 is connected as one, it consists of one cluster, cluster 1. Since node 1 is a seed node, whether or not the seed node is included (seed_yn) is 1, and the cluster size (cluster_size) is 6, which is the number of all included nodes.

Figure 3 shows an example in which a network structure is reconfigured when a failure event is received.

In (a) of FIG. 3, when a failure occurs in the link between node 4 and node 5, the node/link event management unit 30 receives the error and transmits it to the network structure management unit 20 and the network clustering unit 40.

The network structure management unit 20 updates the node information DB 22 and the link information DB 24 according to the received failure information.

The following Tables 4 and 5 show examples of the updated node information DB 22 and link information DB 24.

node_no node_description link_degree cluster_no seed_yn One A 2 One One 2 B One 2 0 3 C 2 One 0 4 D 2 One 0 5 E One 2 0 6 F 2 2 0

node_no connected_node One [3,4] 2 [6] 3 [1,4] 4 [1,3] 5 [6] 6 [2,5]

In Figure 3(a), a failure occurred in the link between node 4 and node 5, and accordingly, a change occurred in node information and link information.

The number of links between Node 4 and Node 5, shown in bold in Table 4, has changed, and therefore the list of connected nodes between Node 4 and Node 5, shown in bold in Table 5, has also changed.

Specifically, node 1, node 4, and node 4 were changed to form the first cluster 101, and node 2, node 5, and node 6 were changed to form the second cluster 102.

The first cluster 101 still includes node 1, which is the seed node, despite the link failure, and thus is not affected by the network.

However, since the second cluster 102 is disconnected from node 1, which is the seed node, it may be determined that a network failure has occurred.

The network clustering unit 40 can update the cluster information DB 42 as shown in Table 6 below.

cluster_no seed_yn cluster_size One One 3 2 0 3

There are two clusters and a cluster number has been added, and the number of nodes included in each cluster is 3, indicating that the second cluster 102 does not include a seed node.

In this way, the network failure determination unit 50 can determine whether there is a failure in divided cluster units 101 and 102.

As an example, if a cluster 102 that does not include a seed node occurs as shown in (a) of FIG. 3, it may be determined that a failure of the entire network has occurred. On the other hand, even if a link failure occurs and the cluster is separated into the third cluster 103 and the fourth cluster 104, as shown in (b) of FIG. 3, if the second node of the fourth cluster 104 is the seed node, the entire network is affected by the failure. It can be judged that it is not affected.

The network failure determination unit 50 may generate a failure cluster slip when a network failure occurs. The fault cluster slip may include the location of the first fault, the number of nodes included in the fault cluster, and the node list.

In (a) of FIG. 3, a failure occurred in the second cluster 102 due to a link failure between node 4 and node 5, so the failure cluster slip includes the link failure between node 4 and node 5, the number of nodes including the failure cluster: 3, The node list Node 2, Node 5, and Node 6 are recorded.

When a node/link failure occurs, the network failure determination unit 50 may merge the failure into the existing failure cluster slip without creating a new fault cluster slip if the failure of the node or link is included in the faults that have already occurred.

Figure 4 shows an example of merging a new event into an existing fault cluster slip.

If the network is divided into the fifth cluster 201 and the sixth cluster 202 due to a link failure between node 4 and node 5, a fault cluster slip for the sixth cluster 106, which is the fault cluster, is created.

Afterwards, when the node/link event management unit 30 additionally receives a failure of node 2, the network failure determination unit 50 determines whether node 2 is included in the existing failure cluster slip. The existing fault cluster slip is a slip for the sixth cluster 106, and the sixth cluster 106 is composed of node 2, node 5, and node 6, and since node 2 is already included, the network fault determination unit 50 The failure information of node 2 is added to the failure slip for the sixth cluster 106 without creating a new slip.

Figure 5 is a flowchart of a network management method according to another preferred embodiment of the present invention. The network management method performed by the network management device of the present invention is summarized once again.

First, node/link events are collected (S10) to determine whether a failure event has occurred (S20), and if a failure does not occur, events are continued to be collected.

If a node/link failure event occurs, it is determined whether the node or link failure is included in the already created failure cluster slip (S30).

If a new failure occurs, the entire network structure is re-clustered due to the node or link failure (S40).

After clustering, it is determined whether a seed node is included in each unit cluster, and if a cluster that does not contain a seed node is created, it is judged to be a faulty cluster and a faulty cluster slip is issued. The fault cluster slip may include the location of the first fault, the number of nodes included in the fault cluster, and the node list.

On the other hand, as a result of determining whether the node or link failure is included in the already created fault cluster slip (S30), if the node or link failure is included in the already created fault cluster slip, the node or link failure information is merged into the existing fault cluster slip. (S60).

According to the network management device and method of the present invention as described above, it is possible to quickly determine the impact on the entire network by distinguishing and determining network failure information by cluster rather than by individual node or link.

The scope of protection of the present invention is not limited to the description and expression of the embodiments explicitly described above. In addition, it is to be added once again that the scope of protection of the present invention may not be limited due to changes or substitutions that are obvious in the technical field to which the present invention pertains.

Claims (8)

A network management device comprising one or more processors and memory, wherein the processor:
a network structure management unit that determines the network structure based on node information and link information connecting nodes;
Node/link event management unit that collects failure events of nodes or links;
a network clustering unit that reorganizes the network structure in clusters based on error information generated when a node or link failure occurs; and
a network failure determination unit that determines whether the entire network has a failure in the reconfigured cluster unit;
Including, network management device.
According to paragraph 1,
The network management device is characterized in that the network failure determination unit determines whether the unit cluster has a failure based on whether the reconfigured unit cluster includes a seed node.
According to paragraph 1,
The network management device is characterized in that the network failure determination unit creates a failure cluster slip including a list of nodes and the number of nodes included in a failure cluster among the reconfigured clusters.
According to paragraph 3,
The network failure determination unit merges the failure node or link event into the failure cluster slip if the failure node or link collected by the node/link event management unit is included in the failure cluster slip. Management device.
In a network management method performed by a network management device including one or more processors and memory:
collecting failure events of nodes or links;
Reconfiguring the network structure in clusters based on fault information generated when a node or link failure occurs; and
determining whether the entire network has a failure on a per-reconstructed cluster basis;
Network management method, including.
According to clause 5,
The step of determining whether the entire network has a failure includes determining whether the unit cluster has a failure based on whether the reconfigured unit cluster includes a seed node.
According to clause 5,
The step of determining whether the entire network has a failure is characterized by creating a failure cluster slip including the number of nodes included in the failed cluster among the reconfigured clusters and a node list.
In clause 7,
After collecting the failure event of the node or link, determining whether the node or link in which the failure event occurred is included in the failure cluster slip; further comprising;
A network management method, characterized in that if the node or link where the failure event occurred is included in the failure cluster slip, the fault occurring node or link event is merged into the fault cluster slip.

Images