Network management device and network management method

Abstract

A network management device 1 according to one aspect of the present invention comprises: a storage unit 20 that stores path information indicating elements to be passed through in an internode path from an arbitrary node to another node in a network in a normal state, the network being configured by mesh-connecting a plurality of nodes via links; a path information acquisition unit 11 that acquires path information at the time of failure determination between the plurality of nodes; a path information comparison unit 12 that compares, with respect to each of a plurality of internode sections for which the path information at the time of failure determination has been acquired, a determination path based on the path information at the time of failure determination and a normal path based on the stored path information in a normal state, and identifies a plurality of internode sections of which the determination path is determined to be different from the normal path; and a suspected failure location determination unit 13 that extracts elements that are passed through in the normal path in the identified plurality of internode sections, and determines elements among all the extracted elements that exist in an overlapping manner as suspected failure locations.

Description

Network management apparatus and network management method

[0001] One aspect of the present invention relates to a network management apparatus and a network management method.

[0002] In recent years, various network services using a network constructed of a plurality of nodes, such as communication network services, have been provided. In such a network service, when a failure occurs in a node constituting the network, an operator providing the network service needs to quickly identify the failure location in the network and grasp the influence range of the failure. However, when a failure of a node in the network cannot be autonomously detected, that is, when a so-called silent failure occurs, the operator identifies the failure location in the network, for example, based on a failure report from a user.

[0003] As a method for identifying a failure location in such a case, for example, a method for identifying a failure location in a virtual private network (VPN (Virtual Private Network)), based on failure reports from two or more users, performing an overlay of data communication paths between nodes determined for each user of the users who have reported the failure, and estimating the failure location is known (for example, see Cited Document 1).

[0004] Japanese Patent No. 5495274

[0005] However, the above-described method for identifying a failure location is applicable only when the network to be failure-determined is limited to a VPN and when the data communication path between nodes is uniquely determined for each user, and cannot be applied to a network having a configuration in which there are a plurality of communication paths between nodes for each user.

[0006] For example, in a network that is built over a wide area using a mesh configuration, such as a mobile network or road network, if a failure occurs, multiple paths exist between any given nodes in a mesh network, making it difficult to pinpoint the exact location of the failure simply by overlaying the path information. Furthermore, because a mesh network can continue service by using alternative routes even when a failure occurs, it can be difficult to notice that a failure has occurred.

[0007] This invention has been made in view of the above circumstances, and aims to provide a network management device and network management method that can accurately pinpoint the location of a fault in a network constructed by mesh-connecting multiple nodes.

[0008] A network management device according to one aspect of this invention includes: a storage unit that stores path information indicating elements traversed in the path between any node and another node in a network configured by mesh connection of multiple nodes via links during normal operation; a path information acquisition unit that acquires path information when a failure is detected between multiple nodes; a path information comparison unit that, for each of the multiple nodes from which path information at the time of failure detection has been acquired, compares a determined path based on the path information at the time of failure detection with a normal path based on the stored normal operation path information, and identifies multiple nodes from which the determined path is determined to be different from the normal path; and a failure suspect location determination unit that extracts elements traversed in the normal path between the identified multiple nodes, and determines that any overlapping elements among all extracted elements are suspected failure locations.

[0009] According to one aspect of this invention, it is possible to provide a network management device and a network management method that can accurately pinpoint the location of a fault in a network constructed by mesh-connecting multiple nodes.

[0010] Figure 1 is a schematic diagram showing one example configuration of a network management device according to the embodiment. Figure 2 is a diagram showing one example configuration of a mesh network to be managed by the network management device according to the embodiment. Figure 3 is a diagram showing an example configuration when the mesh network shown in Figure 2 fails. Figure 4 is a graph showing the paths between nodes of the mesh network shown in Figures 2 and 3. Figure 5 is a flowchart showing an example of processing performed by the network management device according to the embodiment. Figure 6 is a diagram showing one example configuration of a mesh network to be managed by the network management device according to the embodiment. Figure 7 is a diagram showing an example configuration when the mesh network shown in Figure 6 fails. Figure 8 is a graph showing the paths between nodes of the mesh network shown in Figures 6 and 7. Figure 9 is a diagram showing an example configuration when the mesh network shown in Figure 6 fails. Figure 10 is a graph showing the paths between nodes of the mesh network shown in Figures 6 and 9.

[0011] Embodiments of this invention will be described below with reference to the drawings.

[0012] Figure 1 is a schematic diagram showing one example configuration of a network management device according to an embodiment. The network management device 1 is, for example, a terminal managed by a service provider that provides network-based services, and manages the network associated with the services provided. In this embodiment, the network managed by the network management device 1 is a network configured by meshing multiple nodes via links (hereinafter sometimes simply referred to as a mesh network). The network managed by the network management device 1 is, for example, a communication network used for mobile communication services, etc. The network management device 1 is configured by, for example, a server computer or a personal computer.

[0013] In this embodiment, the network management device 1 comprises a control unit 10, a storage unit 20, an input unit 30, an output unit 40, and a communication unit 50.

[0014] The input unit 30 acquires various information supplied from outside the network management device 1. The input unit 30 may include, for example, a user interface such as a mouse or keyboard, which is operated by the administrator of the network management device 1 and inputs various information (for example, change information such as numerical values ​​of various settings of the network management device 1, or information indicating operations by the administrator (operation information)), as well as various sensors such as a microphone or touch panel.

[0015] The output unit 40 outputs, for example, visual or auditory information to be presented to the administrator of the network management device 1. The output unit 40 may include, for example, a monitor that visually supplies information to the administrator of the network management device 1. The output unit 40 may also include audio output means, such as a speaker, that audibly supplies information to the administrator of the network management device 1.

[0016] The communication unit 50 includes, for example, one or more wireless communication interface units, enabling the transmission and reception of information with the mesh network managed by the network management device 1. As the wireless interface, for example, an interface employing a low-power wireless data communication standard such as a wireless LAN (Local Area Network) is used.

[0017] The storage unit 20 includes, for example, a main storage unit and an auxiliary storage unit. The main storage unit may include, for example, ROM (read-only memory) and RAM (random-access memory). ROM is a non-volatile memory used exclusively for reading data, and can store data and various setting values ​​used by the control unit 10 in performing various processes. RAM can be used as a so-called work area to temporarily store data when the control unit 10 performs various processes. In this embodiment, the main storage unit is, for example, RAM and is used as memory.

[0018] The auxiliary storage unit is a non-temporary computer-readable storage medium for a computer centered around the control unit 10. Examples of auxiliary storage units include EEPROM® (electric erasable programmable read-only memory), HDD (hard disk drive), or SSD (solid state drive). The auxiliary storage unit can store data used by the control unit 10 in performing various processes, data generated by processing in the control unit 10, or various setting values.

[0019] In this embodiment, the storage unit 20 includes a network configuration information storage unit 21 and a normal route information storage unit 22.

[0020] The network configuration information storage unit 21 stores the configuration information of the network managed by the network management device 1. Specifically, the network configuration information storage unit 21 stores information indicating multiple nodes that constitute a mesh network and multiple links that connect those nodes, as well as information indicating the connection relationships of each node and link. In the following description, nodes and links, which are the unit elements that constitute the network managed by the network management device 1, may be collectively referred to as "elements."

[0021] In addition, in the embodiment, the network configuration information storage unit 21 may store a cost value associated with each element. The cost value is a value that indicates the degree of delay when passing through each element. The cost value is set, for example, based on the delay time of each element in data communication using a communication network. In the following description, the cost value is assumed to be a positive value, and a smaller value is preferable (lower degree of delay).

[0022] The normal route information storage unit 22 stores normal route information for the network managed by the network management device 1. Specifically, the normal route information storage unit 22 stores the normal routes between nodes in a mesh network. Here, a route refers to the path of information or objects between nodes via one or more links. Between nodes refers to the entire route from any node to another node. In other words, route information is information that indicates the elements traversed in the route between nodes.

[0023] In this embodiment, the normal path information storage unit 22 stores normal path information between each node, with an identification flag associated with it. The flag is information used to enable identification between each node in the processing described later. For example, a variable indicating a state where no flag is assigned and a variable indicating a state where a flag is assigned are defined.

[0024] Furthermore, the normal route information storage unit 22 may store the cost value of the route information in association with the normal route information between each node. The cost value of the route information represents the sum of the cost values ​​corresponding to the elements included in each route information.

[0025] The control unit 10 is typically a CPU (Central Processing Unit) and / or a GPU (Graphics Processing Unit), but may also include a processor such as a microcontroller, FPGA (Field Programmable Gate Array), or DSP (Digital Signal Processor). The control unit 10 can realize various functions of the network management device 1 by executing programs such as system software, application software, or firmware stored in the storage unit 20.

[0026] In this embodiment, the control unit 10 includes a mesh network fault location determination processing unit. The mesh network fault location determination processing unit performs a fault location determination process to determine the location of a fault in the network managed by the network management device 1. Specifically, the mesh network fault location determination processing unit includes a route information acquisition unit 11, a route information comparison unit 12, and a suspected fault location determination unit 13.

[0027] The route information acquisition unit 11 acquires current route information between multiple nodes in the network managed by the network management device 1. Specifically, the route information acquisition unit 11 acquires route information at the time of failure detection between multiple nodes in a mesh network.

[0028] The route information comparison unit 12 compares the determined route based on the route information at the time of failure detection with the normal route based on the normal route information stored in the normal route information storage unit 22 for each node from which the route information at the time of failure detection was obtained by the route information acquisition unit 11, and identifies the nodes from which the determined route is determined to be different from the normal route. Here, the determined route is the route between nodes at the time of failure detection, based on the route information at the time of failure detection. The normal route is the normal route between nodes, based on the normal route information.

[0029] The fault suspect location determination unit 13 extracts elements that pass through the normal path between the identified nodes based on the normal path information stored in the normal path information storage unit 22 corresponding to the identified nodes, and determines any overlapping elements among all extracted elements as fault suspect locations. The fault suspect location determination unit 13 also outputs the identified fault suspect locations.

[0030] Next, we will describe the mesh network managed by the network management device 1 according to this embodiment.

[0031] Figure 2 shows an example configuration of a mesh network managed by a network management device according to an embodiment. Network NW1 shown in Figure 2 is a mesh network in which end nodes E1 to E5 and nodes Na to Ne are mesh-connected to each other through links. Here, of the nodes constituting network NW1, only end nodes E1 to E5 can be the starting or ending point in a path using network NW1. For example, if network NW1 is a communication network, end nodes E1 to E4 are endpoint terminals capable of sending or retrieving data, and nodes Na to Ne are relay terminals that transmit received data to other nodes.

[0032] Here, the normal operation of the mesh network is defined as follows: The optimal path is always selected between nodes in the mesh network. The optimal path is, for example, the path with the smallest total cost value between nodes. If there are multiple optimal paths, such as when the total cost values ​​between nodes are the same, one of the multiple optimal paths will be pre-configured. Each node will always use the configured path if it is available.

[0033] For example, in the network NW1 shown in Figure 2, the route E1-Na-E4 is selected as the path between nodes E1-E4 because it has the minimum cost. On the other hand, for the path between nodes E1-E5, the path via node Na (E1-Na-E5) and the path via node Nb (E1-Nb-E5) have the minimum cost, so one of these two paths is set in advance.

[0034] Next, the behavior when a failure occurs within the mesh network is defined as follows:

[0035] Figure 3 shows an example of the configuration in the event of a failure in the mesh network shown in Figure 2. If any of the elements constituting the mesh network fails, the optimal path between nodes will be selected from among the paths that do not pass through the failed element. For example, among the paths that do not pass through the failed element, the path with the minimum cost value will be selected. As in the normal operation, if multiple optimal paths exist, one of the multiple optimal paths will be pre-configured, and only the configured path will be used between nodes.

[0036] For example, as shown in Figure 3, if node Nb of network NW1 fails, the inter-node paths from end node E1 to other end nodes E2 to E5 that pass through node Nb will change.

[0037] Figure 4 is a graph showing the paths between nodes in the mesh network shown in Figures 2 and 3. For example, in network NW1 when a failure occurs as shown in Figure 3, the normal path E1-Nb-E5 between nodes E1 to E5 is unavailable. Therefore, the path between nodes E1 to E5 is selected from among the paths that do not go through node Nb and have the lowest cost value. As shown in Figure 4, the path between nodes E1 to E5 when a failure occurs is selected as the path that does not go through node Nb and has the lowest cost, which is path E1-Na-E5.

[0038] Because a mesh network operates as described above, even if a failure occurs in any part of the network, it is possible to select and use a path that does not pass through the faulty location and has a low degree of delay. For this reason, service providers that offer network services using a mesh network often find it difficult to notice failures in the network and to pinpoint the location of the failure. In order to accurately narrow down the location of the failure even in such a mesh network, the network management device 1 according to this embodiment performs the failure location determination process shown below.

[0039] Figure 5 is a flowchart showing an example of processing performed by the network management device according to the embodiment. First, as a preprocessing step, the control unit 10 of the network management device 1 acquires normal route information for all nodes of the network to be managed and stores it in the normal route information storage unit 22. Alternatively, the control unit 10 of the network management device 1 may calculate the cost value of the normal route information for each node based on the cost value of each element stored in the network configuration information storage unit 21 and store it in association with the route information.

[0040] The control unit 10 of the network management device 1 performs the following fault location determination process based, for example, on a fault report from a user of a network service provided by a business operator managing the network management device 1. Alternatively, the control unit 10 of the network management device 1 performs the following fault location determination process at predetermined intervals. Here, in the following description of the fault occurrence process, "present," which is the time when the fault location determination process is performed, can be read as "time of fault determination," as this is the time when the network management device 1 performs the fault determination process.

[0041] Furthermore, prior to starting the following fault location determination process, the control unit 10 refers to the flags stored in the normal route information storage unit 22 that are associated with the normal route information, and removes any flags that are currently set.

[0042] First, the route information acquisition unit 11 acquires the current route information between all nodes (step S1). The route information acquisition unit 11 acquires the current route information between all nodes in the network to be managed and stores it in a predetermined location in the storage unit 20.

[0043] The route information comparison unit 12 selects an arbitrary pair of nodes (step S2). For example, the route information comparison unit 12 selects one pair of nodes that has not yet been selected from all the pairs of nodes in the network to be managed, and then executes the processes in steps S3 to S5.

[0044] The route information comparison unit 12 determines whether the current route is different from the normal route (step S3). The route information comparison unit 12 compares the determined route based on the acquired current route information with the normal route based on the stored normal route information. When the route information comparison unit 12 determines that the current route is the same as the normal route (step S3: NO), the processing of subsequent steps S4 to S5 is omitted.

[0045] On the other hand, when the route information comparison unit 12 determines that the current route is different from the normal route (step S3: YES), it assigns a flag to the nodes being selected (step S4). The route information comparison unit 12 assigns a flag to the normal route information between the nodes being selected, which is stored in the normal route information storage unit 22.

[0046] The route information comparison unit 12 calculates the cost value of the current route information between the nodes being selected (step S5). The route information comparison unit 12 calculates the total value of the cost values of the elements through which the current route between the nodes being selected passes, based on the cost values of each element stored in the network configuration information storage unit 21.

[0047] The route information comparison unit 12 determines whether all node pairs have been selected (step S6). The route information comparison unit 12 determines, for example, whether all node pairs in the network that is the management target of the network management device 1 in step S2 have been selected. When the route information comparison unit 12 determines that not all node pairs have been selected (step S6: NO), it returns to step S2 and repeats the processing of steps S2 to S5.

[0048] On the other hand, when the route information comparison unit 12 determines that all node pairs have been selected (step S6: YES), the suspected failure location determination unit 13 extracts the elements indicated by the normal route information for the flagged node pairs (step S7). The suspected failure location determination unit 13 identifies the route information with a flag among the normal route information between all node pairs stored in the normal route information storage unit 22. The suspected failure location determination unit 13 extracts all the elements through which the normal route based on the identified normal route information passes, and holds them in a predetermined location in the storage unit 20.

[0049] Alternatively, in step S7, the fault suspect location determination unit 13 may extract all elements indicated by the normal route information for nodes among the flagged nodes where the difference (absolute value of the difference) between the cost value of the normal route information and the cost value of the current route information is greater than or equal to a predetermined threshold. The fault suspect location determination unit 13 stores all extracted elements in a predetermined location in the storage unit 20.

[0050] The fault suspect location determination unit 13 selects an arbitrary element (step S8). The fault suspect location determination unit 13 then selects one element from all the extracted elements that has not yet been selected and executes the processes in steps S9 to S11.

[0051] The fault suspect location determination unit 13 determines whether the selected element exists in duplicate (step S9). The fault suspect location determination unit 13 determines whether the selected element is included multiple times among all extracted elements. That is, the fault suspect location determination unit 13 determines whether the element is duplicated when the normal paths between nodes, which have been determined to have changed paths compared to normal, are superimposed. If the fault suspect location determination unit 13 determines that the selected element does not exist in duplicate (step S9: NO), it omits the processing in steps S10 to S11.

[0052] On the other hand, if the fault suspect location determination unit 13 determines that the selected element is duplicated (step S9: YES), it determines whether or not there is a path that passes through the selected element among the current paths between flagged nodes (step S10). The fault suspect location determination unit 13 determines whether or not there is a determination path that passes through the selected element based on the path information between flagged nodes among the current path information between all acquired nodes. If the fault suspect location determination unit 13 determines that there is a current path that passes through the selected element (step S10: NO), it omits the processing in step S11.

[0053] On the other hand, if the fault suspect location determination unit 13 determines that there is no current path passing through the selected element (step S10: YES), it identifies the selected element as the fault suspect location (step S11). The fault suspect location determination unit 13 stores the selected element in a predetermined location in the storage unit 20.

[0054] The fault suspect location determination unit 13 determines whether all elements have been selected (step S12). The fault suspect location determination unit 13 determines whether all elements extracted in step S8 have been selected. If the fault suspect location determination unit 13 determines that not all elements have been selected (step S12: NO), it returns to step S8 and repeats the processing in steps S8 to S11.

[0055] On the other hand, if the suspected fault location determination unit 13 determines that all elements have been selected (step S12: YES), it outputs the suspected fault location (step S13). The suspected fault location determination unit 13 outputs all elements held in step S11 as suspected fault locations. The suspected fault location determination unit 13 outputs the suspected fault locations to the administrator of the network management device 1, for example, via the output unit 40. Alternatively, the suspected fault location determination unit 13 stores the suspected fault locations in a predetermined location in the storage unit 20 as a fault diagnosis processing result. After completing the processing in step S13, the suspected fault location determination unit 13 terminates the above fault location determination process.

[0056] Next, we will explain the operation of the fault location detection process of the network management device 1 described above, using a specific example.

[0057] Figure 6 shows an example configuration of a mesh network managed by the network management device according to the embodiment. Network NW2 shown in Figure 6 is a mesh network in which end nodes E1 to E6 and nodes Na to Nf are mesh-connected to each other through links. Here, of the nodes constituting network NW2, only end nodes E1 to E6 can be the starting or ending point of a path using network NW2.

[0058] In the following specific example, the network management device 1 is connected to the network NW2 shown in Figure 6, and manages the network NW2. Specifically, the network configuration information storage unit 21 of the network management device 1 stores information indicating the elements that constitute the network NW2, and information indicating the connection relationships of each element. Furthermore, the control unit 10 of the network management device 1 acquires normal path information between each end node of the network NW2 as a preprocessing step for fault location detection and stores it in the normal path information storage unit 22.

[0059] Furthermore, the control unit 10 of the network management device 1 calculates the cost value of the normal route information between each node based on the cost value of each element stored in the network configuration information storage unit 21, and stores it in the normal route information storage unit 22 in association with the route information between nodes.

[0060] Figure 7 shows an example of the configuration in the event of a failure in the mesh network shown in Figure 6. First, as a specific example, the operation of the failure location determination process of the network management device 1 when link E1-Na of network NW2 fails will be explained according to the flowchart shown in Figure 5. The network management device 1 performs the failure location determination process based on a failure report from a user of a network service using network NW2, for example.

[0061] In step S1, the route information acquisition unit 11 acquires current route information for all nodes in network NW2. Next, the route information acquisition unit 11 performs the processes in steps S2 to S6 for all nodes in network NW2 and compares the normal route with the current route.

[0062] Figure 8 is a graph showing the paths between nodes in the mesh network shown in Figures 6 and 7. The graph in Figure 8 shows, as an example, the paths between nodes in network NW2, starting from end node E1.

[0063] As can be seen in Figure 8, the current paths between nodes E1-E4, E1-E5, and E1-E6, starting from end node E1, are different from the normal paths. Therefore, the path information comparison unit 12 assigns a flag to the nodes, including nodes E1-E4, E1-E5, and E1-E6, that it has determined to have different current paths from the normal paths.

[0064] Next, in step S7, the fault suspect location determination unit 13 extracts elements between the flagged nodes based on the normal route information stored in the normal route information storage unit 22.

[0065] The suspected fault location determination unit 13 performs the processing in steps S8 to S12 for all extracted elements to determine the suspected fault location.

[0066] Specifically, in step S9, the fault suspect location determination unit 13 identifies elements that are duplicated among all extracted elements as elements that may be faulty. For example, looking at Figure 8, it can be seen that among the elements traversed in the normal paths between nodes E1-E4, E1-E5, and E1-E6, where the path has been determined to have changed, node Na and link E1-Na are duplicated. The fault suspect location determination unit 13 identifies node Na and link E1-Na as duplicated elements.

[0067] In step S10, the fault suspect location determination unit 13 determines whether there is a path that passes through an element identified as overlapping among the current paths between nodes where the path has been determined to have changed. That is, the fault suspect location determination unit 13 excludes elements that are currently in use from the fault suspect location among the elements identified as potentially faulty locations. For example, looking at Figure 8, among the node Na and link E1-Na identified as fault suspect locations, node Na is used as the current path between nodes E1 to E6, so node Na is excluded from the fault suspect location.

[0068] Subsequently, the fault suspect location determination unit 13 identifies and retains link E1-Na as the fault suspect location in step S11, and outputs it in step S13. In this way, the network management device 1 can identify the fault of link E1-Na in the mesh network NW2 shown in Figure 7.

[0069] Next, the operation of the fault location determination process of the network management device 1 will be explained using other failure examples as concrete examples.

[0070] Figure 9 shows an example of the configuration in the event of a failure in the mesh network shown in Figure 6. Below, as a specific example 2, the operation of the failure location determination process of the network management device 1 when node Na of network NW2 fails will be explained according to the flowchart shown in Figure 5. The network management device 1 performs the failure location determination process based on a failure report from a user of a network service using network NW2, for example.

[0071] In step S1, the route information acquisition unit 11 acquires current route information for all nodes in network NW2. Next, the route information acquisition unit 11 performs the processes in steps S2 to S6 for all nodes in network NW2 and compares the normal route with the current route.

[0072] Figure 10 is a graph showing the paths between nodes in the mesh network shown in Figures 6 and 9. The graph in Figure 10 shows, as an example, the paths between some of the nodes in network NW2 out of all the nodes in the network.

[0073] Figure 10 shows that the current paths between nodes E1-E4, E1-E5, and E1-E6, which start from end node E1, and between nodes E2-E6, which start from end node E2, differ from the normal paths. Therefore, the path information comparison unit 12 assigns a flag to the nodes, including between nodes E1-E4, E1-E5, E1-E6, and E2-E6, that it has determined to have different current paths from the normal paths.

[0074] In this specific example 2, the route information comparison unit 12 calculates the sum of the cost values ​​of the elements traversed in the current route, based on the acquired current route information, for each node between which it has determined that the normal route and the current route are different. The route information comparison unit 12 stores the calculated cost values ​​of the current route information between each node in a predetermined location in the storage unit 20.

[0075] Next, in step S7, the fault suspect location determination unit 13 extracts elements between the nodes that have been flagged, based on the normal path information.

[0076] In this specific example 2, the fault suspect location determination unit 13 calculates the absolute difference between the cost value of the normal route information stored in the normal route information storage unit 22 and the calculated cost value of the current route information for each node to which a flag has been assigned. For nodes to which the calculated absolute value is greater than or equal to a predetermined threshold, the fault suspect location determination unit 13 extracts the elements indicated by the normal route information stored in the normal route information storage unit 22. In this specific example 2, the predetermined threshold is set to "10".

[0077] In Figure 10, it can be seen that among the flagged node pairs E1-E4, E1-E5, E1-E6, and E2-E6, the difference between the cost under normal conditions and the cost during failure is 0 for node pair E1-E4. That is, the fault suspect location determination unit 13 extracts elements indicated by the normal path information stored in the normal path information storage unit 22 for the node pairs including E1-E5, E1-E6, and E2-E6, excluding node pairs E1-E4, etc., where the difference between the cost under normal conditions and the cost during failure is less than 10.

[0078] Subsequently, the suspected fault location determination unit 13 performs the processing in steps S8 to S12 for all extracted elements to determine the suspected fault location.

[0079] Specifically, in step S9, the fault suspect location determination unit 13 identifies elements that overlap with the elements traversed in the normal path between nodes whose paths have changed as elements that may be faulty. For example, looking at Figure 10, it can be seen that among the elements of the normal path between nodes E1-E5, E1-E6, and E2-E6, which are determined to have changed paths, node Na and link E1-Na overlap. The fault suspect location determination unit 13 identifies node Na and link E1-Na as overlapping elements.

[0080] In step S10, the fault suspect location determination unit 13 determines whether there are any paths that pass through elements identified as overlapping among the current paths between nodes where the path has been determined to have changed. That is, among the elements identified as potentially faulty locations, the fault suspect location determination unit 13 excludes currently used elements (where the failure should have occurred) from the fault suspect location. For example, looking at Figure 10, among the current paths between nodes starting from end node E1, there are no paths that pass through node Na and link E1-Na, which have been identified as overlapping. Also, looking at other paths between nodes that pass through node Na as part of the normal path, such as between nodes E2 to E6 shown in Figure 10, there are no current paths that pass through node Na. Therefore, in step S11, the fault suspect location determination unit 13 identifies and retains node Na and link E1-Na as fault suspect locations.

[0081] Subsequently, the fault location determination unit 13 outputs the identified fault location, including node Na and link E1-Na. At this time, the output fault location is likely to include not only node Na, which is the actual fault location, but also other links connected to node Na, including the aforementioned link E1-Na. However, in this case, for example, the administrator of the network management device 1 can easily infer from the output result that node Na, which has been identified as a fault location and to which many other links identified as fault locations are commonly connected, is likely to be the actual fault location.

[0082] As described above, the network management device 1 according to the embodiment includes: a storage unit 20 that stores path information indicating elements traversed in the path between any node and another node in a network configured by mesh connection of multiple nodes via links during normal operation; a path information acquisition unit 11 that acquires path information when a failure is detected between multiple nodes; a path information comparison unit 12 that, for each of the multiple nodes from which path information at the time of failure detection has been acquired, compares the determined path based on the path information at the time of failure detection with the normal path based on the stored normal path information and identifies multiple nodes from which the determined path is determined to be different from the normal path; and a failure suspect location determination unit 13 that extracts elements traversed in the normal path between the identified multiple nodes and determines that any overlapping elements among all extracted elements are suspected failure locations.

[0083] As a result of the configuration described above, the network management device 1 according to this embodiment can accurately pinpoint the location of a fault even within a network constructed by mesh-connecting multiple nodes.

[0084] In the description of the operation of the fault location determination process performed by the control unit 10 of the network management device 1 as described above, the route information acquisition unit 11 acquires route information between all nodes in the network managed by the network management device 1 when a fault is determined. However, it is not necessarily required to acquire route information between all nodes. For example, the route information acquisition unit 11 may be configured to acquire route information between multiple nodes starting from the node used by the user who reported the fault. However, acquiring more route information when determining a fault can improve the accuracy of narrowing down the fault location in the fault location determination process.

[0085] Furthermore, in the network management device 1 according to the embodiment, the storage unit 20 further stores a cost value indicating the degree of delay when passing through each element constituting the network, and the fault suspect location determination unit 13 further calculates the absolute value of the difference between the sum of the cost values ​​of the elements passed through in the normal path and the sum of the cost values ​​of the elements passed through in the determination path for each of the identified plurality of nodes, and extracts elements passed through in the normal path only for nodes where the absolute value is equal to or greater than a predetermined threshold.

[0086] In the fault location determination process performed by the control unit 10 of the network management device 1 according to this embodiment, the process using cost values, as described in, for example, Specific Example 2, is not mandatory, and the process using cost values ​​can be omitted, as described in Specific Example 1.

[0087] However, in a mesh-configured communication network, such as a mobile network, when each communication terminal, which is a node constituting the communication network, performs load balancing, each communication terminal appropriately selects the path with the lowest current latency from among many paths to the data communication endpoint. For this reason, in the above-mentioned communication network, even if no failure occurs within the network, there may be cases where the normal path stored in the memory unit 20 differs from the current path.

[0088] In this case, by using cost values ​​for processing, even if the normal route and the current route differ, if the difference in cost values ​​is small, the process can be excluded and proceed. This makes it possible to perform fault location detection processing while excluding cases where the normal route is continuously used due to load balancing.

[0089] Furthermore, in the network management device 1 according to the embodiment, the fault suspect location determination unit 13 further determines, for each element determined to exist in duplicate, whether or not that element is traversed in the determination path between the identified plurality of nodes, and excludes the element determined to be traversed from the fault suspect location.

[0090] As described above, when the normal path is superimposed on the path used during fault detection between nodes where the path has been determined to have changed compared to the normal path, and overlapping elements are extracted as potential fault locations, it is possible that not only the actual fault location but also other elements connected to the fault location may be extracted as potential fault locations. The network management device 1 according to the embodiment can exclude other elements that are not the actual fault location from the output result by determining whether or not the overlapping elements are used in the path at the time of fault detection.

[0091] It should be noted that the present invention is not limited to the embodiments described above, and can be modified in various ways during implementation without departing from its essence. Furthermore, each embodiment may be combined as appropriate, and in that case, the combined effects can be obtained. Moreover, the above embodiments include various inventions, and various inventions can be extracted by selecting combinations from the multiple constituent elements disclosed. For example, if the problem can be solved and effects obtained even if some constituent elements are deleted from all the constituent elements shown in the embodiment, then the configuration with these deleted constituent elements can be extracted as an invention.

[0092] 1...Network management device 10...Control unit 11...Route information acquisition unit 12...Route information comparison unit 13...Fault suspect location determination unit 20...Storage unit 21...Network configuration information storage unit 22...Normal route information storage unit 30...Input unit 40...Output unit 50...Communication unit NW1, NW2...Network

Claims

1. A network management device comprising: a storage unit that stores path information indicating elements traversed in the path between any node and another node in a network configured by mesh connection of multiple nodes via links during normal operation; a path information acquisition unit that acquires path information when a failure is detected between multiple nodes; a path information comparison unit that, for each of the multiple nodes from which path information at the time of failure has been acquired, compares a determined path based on the path information at the time of failure with a normal path based on the stored normal operation path information, and identifies multiple nodes from which the determined path is determined to be different from the normal path; and a failure suspect location determination unit that extracts the elements traversed in the normal path between the identified multiple nodes, and determines that any overlapping elements among all extracted elements are suspected failure locations.

2. The network management device according to claim 1, wherein the storage unit further stores a cost value indicating the degree of delay when passing through each of the elements constituting the network, and the fault suspect location determination unit further calculates, for each of the identified plurality of nodes, the absolute value of the difference between the sum of the cost values ​​of the elements passed through in the normal path and the sum of the cost values ​​of the elements passed through in the determination path, and extracts the elements passed through in the normal path only for the nodes where the absolute value is equal to or greater than a predetermined threshold.

3. The network management device according to claim 1, wherein the fault suspect location determination unit further determines, for each of the elements determined to exist in duplicate, whether or not that element is traversed in the determination path between the identified plurality of nodes, and excludes the element determined to be traversed from the fault suspect location.

4. A network management method performed by the control unit of a network management device, comprising: acquiring path information indicating elements traversed in the path between any node in a network configured by mesh connection of multiple nodes via links at the time of fault detection; comparing a determined path based on the path information at the time of fault detection with a normal path indicating the normal path between the nodes for each of the multiple nodes from which path information at the time of fault detection was acquired; identifying multiple nodes from which the determined path is determined to be different from the normal path; extracting the elements traversed in the normal path between the identified multiple nodes; and determining any overlapping elements among all extracted elements as suspected fault locations.

Images