JP2017108300A - Link failure pattern counting system, link failure pattern counting method, and link failure pattern counting program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To count efficiently the number of failure patterns for which recalculation of a route is required for every number of link failures.SOLUTION: A deletion part 12, in a network topology in which nodes are connected by links, generates a topology in a case where the links are deleted one by one; and when all nodes have links in any of the generated topology, with respect to the topology in which all nodes have the links, generates topology in a case where the links are deleted one by one, while defining a plurality of topologies in which connection relations of the links connected to the nodes are similar as topology of the same kind. Further, a pattern addition part 13, with respect to topology in which nodes having no links exist of the topology generated by the deletion part 12, sums up the number of link failure patterns in which the topology can be generated for each number of deletion of the links deleted until the topology is generated.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a link failure pattern number calculation system, a link failure pattern number calculation method, and a link failure pattern number calculation program.

  For example, when the topology of a network composed of a plurality of routers changes due to a failure or the like, the router may need to recalculate a route to transfer a packet based on the latest topology. At this time, since the loop and packet drop occur until the route information of all the routers is updated, the communication quality is greatly deteriorated.

  Thus, FastRoute is known as a technique for reducing the route update time by switching the route to a preset Multi-Protocol Label Switching (MPLS) tunnel when the route is recalculated.

  Furthermore, as a technique for solving the problem of FastRoute that it is necessary to prepare a huge number of tunnels to cope with all failure patterns, packet transfer is continued only by switching the next hop. A method is known (for example, Non-Patent Document 1).

  In the method of Non-Patent Document 1, when a physical topology changes due to a link failure or the like, a logical topology created based on the physical topology is used to determine whether or not packet transfer can be continued by switching the next hop (NH). Judgment. At this time, the physical topology changes, but the logical topology does not change, and if it is determined that the packet transfer can be continued by switching the NH, each router continues the packet transfer by switching the NH. On the other hand, when a change occurs in the logical topology, the router recalculates the route.

  In this way, when it is necessary to recalculate the route and recalculate as necessary, when redesigning the network, the path recalculation occurrence probability for each number of failed links and the number of failed links It is important to know the average number of link failures calculated based on the path recalculation probability. Here, the probability of occurrence of path recalculation for each number of faulty links is obtained by calculating the number of fault patterns that require path recalculation for each number of link faults with respect to the total number of fault patterns.

  Specifically, the following methods are known as methods for calculating the number of failure patterns that require route recalculation. First, the configuration of a physical link connecting adjacent logical nodes is acquired from the logical network topology of the actual network topology. Next, the physical links are deleted in a predetermined order, and the number of link deletions when the route needs to be recalculated is measured. Then, by performing these procedures for all possible orders, the number of failure patterns that require route recalculation for each number of link failures is obtained.

Koji Sugizono, Aki Fukuoka, Hiroshi Yamazaki, Taso Yamamoto "Proposal on packet transfer method considering avoidance of route recalculation" IEICE National Convention March 2015

  However, the conventional method has a problem that it is not possible to efficiently calculate the number of failure patterns that require route recalculation for each number of link failures.

  For example, when calculating the number of failure patterns that require route recalculation for each number of link failures in the conventional method, the calculation time increases rapidly as the number of links increases, and the calculation efficiency also deteriorates rapidly. . This is because all failure patterns exist by the factorial of the number of links.

  The link failure pattern number calculation system of the present invention generates a topology when each link is deleted one by one in a topology of a network in which nodes are connected by links, and in any of the generated topologies, If all nodes have links, delete all links with the same type of topologies that have the same connection relationship of the links connected to the nodes. For each topology that has been deleted before the generation of the topology, there is a deletion unit that generates the topology in each case, and a topology that includes a node that does not have a link among the topologies generated by the deletion unit, The total number of patterns for deleting links that can generate the topology It characterized by having a a calculation unit.

  Also, the link failure pattern number calculation method of the present invention generates a topology when each link is deleted one by one in a topology of a network in which nodes are connected by links, and any one of the generated topologies In the case where all nodes have links, with respect to the topology in which all the nodes have links, a plurality of topologies having the same connection relation of links connected to the nodes are regarded as the same type topology, and each link has one link. A deletion step for generating topologies in the case of each deletion, and for each topology deleted by the generation of the topology, among the topologies generated by the deletion step, there is a topology in which a node having no link exists. The total number of link deletion patterns that can generate the topology Characterized in that it includes a down count adding step.

  According to the present invention, it is possible to efficiently calculate the number of failure patterns that require route recalculation for each number of link failures.

FIG. 1 is a diagram for explaining a network. FIG. 2 is a diagram for explaining a link failure. FIG. 3 is a diagram for explaining the number of all link failure patterns. FIG. 4 is a diagram illustrating an example of the configuration of the link failure pattern number calculation system according to the first embodiment. FIG. 5 is a diagram for explaining failure patterns having the same topology. FIG. 6 is a diagram for explaining data stored in the pattern storage unit. FIG. 7 is a diagram for explaining failure patterns having different topologies. FIG. 8 is a diagram for explaining the topology G1. FIG. 9 is a diagram for explaining the topology G2. FIG. 10 is a diagram for explaining data stored in the pattern storage unit. FIG. 11 is a diagram for explaining a failure pattern that requires recalculation of a route. FIG. 12 is a diagram for explaining data stored in the pattern storage unit. FIG. 13 is a flowchart showing the flow of processing of the link failure pattern number calculation system according to the first embodiment. FIG. 14 is a diagram for explaining the second embodiment. FIG. 15 is a diagram for explaining the second embodiment. FIG. 16 is a diagram for explaining the third embodiment. FIG. 17 is a flowchart showing the flow of processing of the link failure pattern number calculation system according to the third embodiment. FIG. 18 is a diagram illustrating an example of a computer that realizes a link failure pattern number calculation system by executing a program.

  Hereinafter, an embodiment of a link failure pattern number calculation system according to the present application will be described in detail with reference to the drawings. In addition, this invention is not limited by this embodiment. Moreover, each embodiment can be combined suitably.

[First Embodiment]
First, the configuration, process flow, and effects of the policy control system and policy control device according to the first embodiment will be described.

  First, a failure of a network and a link targeted in this embodiment will be described. First, as shown in FIG. 1, the logical topology of the network 100 includes logical nodes (LN) that are a set of routers and logical links. FIG. 1 is a diagram for explaining a network.

  Here, when a link exists between routers, that is, when routers are adjacent to each other, a logical link exists between logical nodes to which each router belongs. At this time, these logical nodes are said to be adjacent.

Here, each router has the following properties.
-All the logical nodes adjacent to the logical node to which the router belongs have an adjacent router.
-The logical node that is not adjacent to the logical node to which the router belongs does not have an adjacent router.
(However, routers belonging to the same logical node are allowed to be adjacent to each other.)

  As shown in FIG. 1, the routers 111 to 113 belonging to the logical node 110 have links to routers belonging to the logical nodes 120 and 130. Further, the routers 121 to 123 belonging to the logical node 120 have links to routers belonging to the logical node 110. Further, the routers 131 to 133 belonging to the logical node 130 have links to routers belonging to the logical node 110.

  At this time, for example, since a link exists between the router 111 and the router 121, it can be said that the router 111 and the router 121 are adjacent to each other, and the logical node 110 and the logical node 120 are also adjacent. Similarly, it can be said that the logical node 110 and the logical node 130 are adjacent to each other.

  Here, as shown in FIG. 2, for example, when the link between the router 111 and the router 121 fails, the logical node 110 and the logical node 120 are not adjacent to each other, and a route recalculation is required. FIG. 2 is a diagram for explaining a link failure. In this way, the probability of the number of link failures that cause recalculation of the route by focusing on the link connection status between specific adjacent logical nodes and focusing on the link failure occurrence pattern that causes the logical topology conditions to collapse. Can be calculated.

  However, since such a failure occurrence pattern has only the factorial of the number of links as shown in FIG. 3, for example, it may take an enormous calculation time. FIG. 3 is a diagram for explaining the number of all link failure patterns. For example, as shown in FIG. 3, if the number of links is 9, 9! There are street failure patterns. Therefore, in this embodiment, the number of failure patterns that require route recalculation for each link failure number is calculated without paying attention to all patterns.

[Configuration of First Embodiment]
First, the configuration of the link failure pattern number calculation system according to the first embodiment will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of the configuration of the link failure pattern number calculation system according to the first embodiment.

  As illustrated in FIG. 4, the link failure pattern number calculation system 1 includes a control unit 10 and a storage unit 20. The control unit 10 includes an acquisition unit 11, a deletion unit 12, a pattern number addition unit 13, and a pattern update unit 14. The storage unit 20 includes a pattern storage unit 21 and a pattern number storage unit 22.

  First, as shown in FIG. 5, in the logical node 110 in which the routers 111 to 113 having three links to adjacent logical nodes exist, if each router is treated as equivalent, one link from the router 111 is present. In the case where one of the links from the router 112 is disconnected or one link from the router 113 is disconnected, the network is in the same state. FIG. 5 is a diagram for explaining failure patterns having the same topology.

  At this time, the logical node 110 has two routers having three links to adjacent logical nodes and one router having two links to adjacent logical nodes. From this, it can be said that it is not always necessary to consider all occurrence orders when calculating the number of failure patterns that require route recalculation as in the example of FIG.

  The calculation of the number of failure patterns by the link failure pattern number calculation system 1 will be described with reference to FIGS. First, the link failure pattern number calculation system 1 stores data as shown in FIG. 6 in the pattern storage unit 21. FIG. 6 is a diagram for explaining data stored in the pattern storage unit.

  Specifically, first, the acquisition unit 11 acquires a topology that is in an initial state. In the example of FIG. 6, in the topology in the initial state, there are three nodes belonging to the input logical node and three nodes belonging to the output logical node, and the nodes belonging to the input logical node and the output logical node A full mesh connection is established between the nodes to which it belongs. At this time, the number of topology failure patterns in the initial state is 1, and the number of failure links is 0.

  The number of topology failure patterns is the number of link failure patterns that can become the topology as a result of link failure. Further, the number of failed links is the number of links that have failed until the topology is reached. In addition, data in which the topology, the number of topology failure patterns, and the number of failure links are set is stored in the pattern storage unit 21.

  The deletion unit 12 generates a topology when each link is deleted one by one in a network topology in which nodes are connected by links. In addition, in any of the generated topologies, when all nodes have links, for the topologies in which all the nodes have links, a plurality of topologies having the same connection relation of links connected to the nodes are the same type. Since the topology is stored in the pattern storage unit 21, the deletion unit 12 generates a topology when each link is deleted one by one with such a topology as the same type topology.

  As described above, the deletion unit 12 calculates the topology generated as a result of the link failure by deleting the link in the initial topology. At this time, since the result differs depending on the failed link, the deletion unit 12 calculates the topology when each link is deleted. In the example of FIG. 6, since there are nine links in the topology in the initial state, the deletion unit 12 calculates nine types of topologies. The deletion unit 12 deletes the topology data for which the calculation has been completed from the pattern storage unit 21. Here, the deletion unit 12 deletes the topology data in the initial state from the pattern storage unit 21.

  However, these nine types of topologies are the same as described with reference to FIG. That is, if the nodes in the topology in the initial state are numbered and the node numbers of the topology generated as a result of the link failure are exchanged, all become the same topology. Therefore, when a link failure occurs from the initial state, the topology is always the same.

  In the following description, it is assumed that topologies that have the same topology by switching the node numbers are the same topology.

  Further, the pattern storage unit 21 stores, for each topology, the number of failure patterns that is the number of link deletion patterns that can generate the topology. Then, each time the topology is generated by the deletion unit 12, the pattern update unit 14 determines the number of topologies of the same type in which all the nodes have links, and the deletion unit 12 stored in the pattern storage unit 21 sets the links. When the generated topology is stored in the pattern storage unit 21 by multiplying the number of failure patterns of the topology before deletion, the multiplied number is added to the number of failure patterns of the generated topology, When the generated topology is not stored in the pattern storage unit 21, the multiplied number is stored in the pattern storage unit 21 as the number of failure patterns of the generated topology.

  Specifically, the pattern update unit 14 stores the number of failure patterns of the topology G0 obtained here as 9 and the number of failure links as 1 in the pattern storage unit 21. Note that, since the topology G0 is not a pattern that requires recalculation of the route, the pattern number adding unit 13 does not perform processing.

  In the case of FIG. 6, only the topology G0 is generated from the topology in the initial state, but a plurality of different topologies may be generated from one topology as shown in FIG. FIG. 7 is a diagram for explaining failure patterns having different topologies.

  The acquisition unit 11 acquires data having the smallest number of failed links from the pattern storage unit 21. At this time, since only G0 data is stored in the pattern storage unit 21, the acquisition unit 11 acquires G0. Note that when there are a plurality of pieces of data with the smallest number of failed links in the pattern storage unit 21, the acquisition unit 11 acquires arbitrary data.

  Then, the deletion unit 12 deletes the link of the topology G0 and calculates the topology. At this time, the deletion unit 12 obtains the topologies G1 and G2 as a result of the calculation. There are four link failure patterns from topology G0 to topologies G1 and G2. Note that the topologies G1 and G2 are not patterns that require route recalculation, so the pattern number adding unit 13 does not perform processing.

  The topology G1 will be described with reference to FIG. FIG. 8 is a diagram for explaining the topology G1. First, in FIG. 8, a number is assigned to each node and the position of the node is changed for explanation. In FIG. 8, the pattern which becomes topology G1 is a case where any of the links 1-6, 2-6, 3-4 and 3-5 fails. For example, a link between the nodes 1 and 6 is represented as a link 1-6.

  In FIG. 8, the topology generated as a result of the failure of the link 1-6 at the bottom is shown as an example. Let the topology shown at the bottom be G1. Here, if the numbers of the nodes 1 and 2 of the topology generated when the link 2-6 fails, the topology matches the topology G1. When the nodes 3 and 6, the nodes 2 and 5, and the nodes 1 and 4 in the topology generated when the link 3-4 fails, the topology matches the topology G 1. If the nodes 3 and 6, the nodes 1 and 5, and the nodes 2 and 4 of the topology generated when the link 3-5 fails, the topology matches the topology G 1. From this, it can be seen that there are four link failure patterns that result in topology G1 as a result of the link failure.

  At this time, the pattern update unit 14 confirms whether data having the same topology as the topology after link deletion is stored in the pattern storage unit 21. If not stored, the pattern update unit 14 stores the data about the topology in the pattern storage unit. Save to 21. If stored, the number of failure patterns of the stored data is added.

  In the example of FIG. 8, first, when the topology G1 in which the link 1-6 has failed is calculated by the deletion unit 12, the pattern update unit 14 sets the topology to G1, the number of failure patterns, and the number of failed links to 2. Data is stored in the pattern storage unit 21. Next, when the topology G1 in which the link 1-5 has failed is calculated by the deletion unit 12, the pattern update unit 14 adds the failure pattern number 9 to the failure pattern number of the stored data.

  As shown in FIG. 10, when the same processing is repeated for each failure pattern, the number of failure patterns of topology G1 stored in the pattern storage unit 21 is 36. Also, as shown in FIG. 9, the same processing as in the case of topology G1 is performed for topology G2. FIG. 9 is a diagram for explaining the topology G2.

  The acquisition unit 11 further acquires the topology G1 data from the pattern storage unit 21. Then, the deletion unit 12 deletes the link. At this time, as shown in FIG. 11, when a link is deleted from the topology G1, a node with the number of output links being 0 may occur. Such a failure pattern is a failure pattern that requires route recalculation.

  Further, the pattern number adding unit 13 includes, for each topology generated by the deleting unit 12, a number of deleted links, that is, the number of failed links, for a topology in which a node having no link exists. The total number of link failure patterns that can generate the topology is stored.

  Specifically, the pattern number adding unit 13 stores the number of failure links 3 and the number of failure patterns 36 in association with each other in the pattern number storage unit 22. Further, the pattern number adding unit 13 adds 36 to the number of failure patterns when the number of failure patterns corresponding to the number of failure links 3 is already stored in the pattern number storage unit 22. The pattern number storage unit 22 stores the number of failure links and the number of failure patterns, for example, as an array. In this case, the pattern number storage unit 22 stores, for example, A [3] = 36 with the failure pattern number 36 corresponding to the failure link number 3 as an element of the array A.

  As illustrated in FIG. 12, the pattern update unit 14 adds the number of failure patterns calculated in the process for the topology G2 to the data in the pattern storage unit 21 when the processes are performed in the order of topology G1 → G2. FIG. 12 is a diagram for explaining data stored in the pattern storage unit.

  In the example of FIG. 12, there are four link failure patterns that result in topology G3 as a result of link failure in topology G1. Further, since the number of link failure patterns that become topology G1 from the initial state is 36, the number of failure patterns that become G3 is 144 multiplied by 4. The number of failed links from the initial topology to the topology G3 is 3.

  At the time when the processing related to the topology G1 is completed, the topology G2 and the topology G3 exist in the pattern storage unit 21. Of the failure patterns of the links in topology G2, there are two failure patterns that become topology G3. Here, since the data of the topology G3 has already been stored in the pattern storage unit 21, 72 obtained by multiplying 36 by 2 is added to the number 144 of stored failure patterns.

  Note that the number of link failure patterns for each number of failed links is required when calculating the average number of failed links when there is a case where recalculation of the route is required. In that case, it is necessary to calculate the recalculation occurrence probability for each number of failed links. The calculation method of the recalculation occurrence probability for each number of failed links is as follows.

  First, a value obtained by multiplying the number of failure patterns calculated in the present embodiment by the factor of the number of remaining links, which is the number of link failure patterns when the number of remaining links is broken, is calculated. Next, the calculated value is divided by the factorial value of the number of links in the initial state, which is the total number of link failure patterns resulting from the topology in the initial state. The value obtained at this time is the recalculation occurrence probability for each number of failed links.

  However, the recalculation occurrence probability for each number of failed links calculated here is based on the premise that the failure probability of all links is the same. Then, multiply the number of failed links by the number of failed links and the recalculation probability for each number of failed links, and calculate the average number of failed links when path recalculation occurs by taking the sum of the obtained values. can do.

[Process of First Embodiment]
The processing of the first embodiment will be described with reference to FIG. FIG. 13 is a flowchart showing the flow of processing of the link failure pattern number calculation system according to the first embodiment. First, as illustrated in FIG. 13, the acquisition unit 11 acquires the topology in the initial state, and inputs the acquired topology into the function F (step S101). Here, a series of processing is performed by the function F.

  Next, the pattern update unit 14 registers data obtained by combining the input topology, the number of failure links 0, and the number of failure patterns 1 in the pattern storage unit 21 (step S102). Then, when there is data in the pattern storage unit 21 (step S103, Yes), the acquisition unit 11 takes out the data from the pattern storage unit 21 and sets it as a processing target (step S104). If there is no data in the pattern storage unit 21 (No at Step S103), the process ends.

  The deletion unit 12 deletes one topology link included in the data, and generates a topology generated when one link fails (step S105). Then, when the same topology exists among the generated topologies, the pattern update unit 14 counts the link failure patterns that become the topology (step S106).

  Here, when data including the same topology as the generated topology exists in the pattern storage unit 21 (step S107, Yes), the pattern update unit 14 counts the counted pattern as the number of failure patterns of the data including the same topology. The number multiplied by the number of link failure patterns included in the data output from the pattern storage unit 21 is added (step S109).

  Here, in the case where data including the same topology as the generated topology does not exist in the pattern storage unit 21 (No in step S107), there is a node in which the number of output links is 0 in the generated topology (step S107). S108, Yes), the pattern number adding unit 13 sets the value of the data link failure pattern number to the value of the data failure link number in the array for counting the number of failure patterns stored in the pattern number storage unit 22. Add together (step S111). If there is no node with the number of output links 0 in the generated topology (No in step S108), the process proceeds to the next process.

  Finally, the pattern update unit 14 calculates the number obtained by multiplying the generated topology by the topology, the number of counted patterns, and the number of link failure patterns included in the data output from the pattern storage unit 21, and the number of failure patterns. Data in which the number obtained by adding 1 to the number is the number of failed links is stored in the pattern storage unit 21 (step S110). And it returns to step S103 and repeats a process.

[Effect of the first embodiment]
The deletion unit 12 generates a topology when each link is deleted one by one in a network topology in which nodes are connected by links, and all nodes in the generated topology have links to each other. In the case of having a topology in which all the nodes have links, a topology in which a plurality of topologies having the same connection relation of links connected to the nodes are regarded as the same type topology and each link is deleted one by one. Generate each.

  The pattern number adding unit 13 generates a topology for each of the deleted links that have been deleted before the generation of the topology, among the topologies generated by the deleting unit 12, in which there is a node having no link. Sum the number of possible link failure patterns.

  As described above, in the first embodiment, a plurality of topologies having the same connection relation of links connected to nodes are regarded as the same type of topology. Therefore, it is not necessary to consider the necessity of recalculation for all the failure patterns, and it is possible to efficiently calculate the number of failure patterns that require route recalculation for each number of link failures.

  Further, the pattern storage unit 21 stores, for each topology, the number of failure patterns that is the number of failure patterns of links that can generate the topology. Then, each time the topology is generated by the deletion unit 12, the pattern update unit 14 determines the number of topologies of the same type in which all the nodes have links, and the deletion unit 12 stored in the pattern storage unit 21 sets the links. When the generated topology is stored in the pattern storage unit 21 by multiplying the number of failure patterns of the topology before deletion, the multiplied number is added to the number of failure patterns of the generated topology, When the generated topology is not stored in the pattern storage unit 21, the multiplied number is stored in the pattern storage unit 21 as the number of failure patterns of the generated topology. Thereby, the pattern number adding unit 13 can calculate the number of failure patterns for each number of failure links based on the number of failure patterns acquired from the pattern storage unit 21.

[Second Embodiment]
In the first embodiment, the number of failure patterns is stored in the pattern storage unit 21. In contrast, in the second embodiment, the recalculation probability calculation is simplified by storing the occurrence rate of each pattern in the pattern storage unit 21 as a failure pattern ratio.

  In the second embodiment, the pattern storage unit 21 can generate, for each topology, the number of failure patterns of the links that can generate the topology before the topology is deleted by the deletion unit 12. A failure pattern ratio that is a ratio to the number of failure patterns is further stored. In addition, each time the topology is generated by the deletion unit 12, the pattern update unit 14 determines the number of the same type of topology in which all the nodes have links by the deletion unit 12 stored in the pattern storage unit 21. When the number divided by the number of failure patterns of the topology before deletion is multiplied by the failure pattern ratio of the topology before the link is deleted by the deletion unit 12, and the generated topology is stored in the pattern storage unit 21 The multiplied number is added to the failure pattern rate of the generated topology, and when the generated topology is not stored in the pattern storage unit 21, the multiplied number is added to the failure pattern of the generated topology. It is stored in the pattern storage unit 21 as a ratio.

  As illustrated in FIG. 14, the pattern update unit 14 stores the failure pattern ratio in the pattern storage unit 21. FIG. 14 is a diagram for explaining the second embodiment. As shown in FIG. 14, the failure pattern ratio is obtained, for example, by dividing the number of link failure patterns 9 that results in topology G0 by the number 9 of links in the initial topology. When the number of failure links is 1, all failure patterns are G0, so the failure pattern ratio is 1.

  Further, the topologies G1 and G2 are as shown in FIG. The number of link failure patterns that change from topology G0 to G1 and G2 is 4 patterns as described above. As shown in FIG. 15, the failure pattern ratio 1 of G0 is multiplied by 1/2, which is the number of link failure patterns 4 that change from topology G0 to topology G1 divided by 8 that is the number of remaining links of topology G0. Thus, the failure pattern ratio of G1 is obtained. At this time, the failure pattern ratio of G1 is ½.

[Effects of Second Embodiment]
For each topology, the pattern storage unit 21 has a ratio of the number of failure patterns of the link that can generate the topology to the number of failure patterns of the link that can generate the topology before the topology is deleted by the deletion unit 12. A certain failure pattern ratio is further stored. In addition, each time the topology is generated by the deletion unit 12, the pattern update unit 14 determines the number of the same type of topology in which all the nodes have links by the deletion unit 12 stored in the pattern storage unit 21. When the number divided by the number of failure patterns of the topology before deletion is multiplied by the failure pattern ratio of the topology before the link is deleted by the deletion unit 12, and the generated topology is stored in the pattern storage unit 21 The multiplied number is added to the failure pattern rate of the generated topology, and when the generated topology is not stored in the pattern storage unit 21, the multiplied number is added to the failure pattern of the generated topology. It is stored in the pattern storage unit 21 as a ratio. Thereby, when calculating the recalculation occurrence probability, the failure pattern ratio can be used, so that the calculation amount can be reduced.

[Third Embodiment]
In the first embodiment, among the topologies stored in the pattern storage unit 21, the processing is performed in the order of the number of fault links, but in the third embodiment, the topology is called by calling a recursive function. Each time is generated, processing is performed on the generated topology.

  A third embodiment will be described with reference to FIG. FIG. 16 is a diagram for explaining the third embodiment. As shown in FIG. 16, in the third embodiment, after a link is deleted from the topology processed by the function and the topology is generated, there is a topology having the same number of failed links as the topology being processed. Even in this case, the previously generated topology is input to the function for processing.

  First, the initial topology is input to the recursive function F. The recursive function F calculates all the topologies caused by link failures from the initial state, and calculates the number of link failure patterns that change to each topology. As described above, when a link is deleted from the initial state, topology G0 is obtained regardless of which link is deleted.

  Next, for all the calculated topologies, each topology and the number of link failure patterns of each topology are input to the function F. Here, since the calculated topology is only G0, the topology G0 and 9 which is the number of link failure patterns that change from the initial state to the topology G0 are input to the recursive function F.

  Next, the recursive function F performs processing when the link of the topology G0 fails. As described above, topologies G1 and G2 are generated from topology G0. Here, the number of link failure patterns that become G1 and the number of failure patterns that become G2 are both 4. Therefore, the number obtained by multiplying 4 by the number of link failure patterns input to the recursive function F, that is, 4 × 9 = 36 is the number of failure patterns of the topologies G1 and G2.

  Here, the topologies G1 and G2 are sequentially input to the recursive function F. Note that while the recursive function F is executed with the topology G1 as an input, the processing of the recursive function F with the topologies G3, G4, and G5 generated from the topology G1 as inputs is executed. After all these processes are completed, the process of the recursive function F that receives the topology G1 is completed. Thereafter, processing of the recursive function F having the topology G2 as an input is executed.

  In the first embodiment, the processing is performed in the order of G0 → G1 → G2 → G3 → G4... Whereas in the third embodiment, the processing is performed in the order of G0 → G1 → G3 → G4. Processing is performed. That is, in the first embodiment, processing is performed in order from the topology having the smallest number of failed links, whereas in the third embodiment, processing is performed in order from the topology having the largest number of failed links.

  In this way, the deletion unit 12 may delete links in order from the topology with the largest number of failed links out of the input topologies, or may delete links in order from the topology with the smallest number of failed links. .

[Process of Third Embodiment]
The process of the third embodiment will be described with reference to FIG. FIG. 17 is a flowchart showing the flow of processing of the link failure pattern number calculation system according to the third embodiment. As shown in FIG. 17, first, the acquisition unit 11 inputs the topology in the initial state and the number of failure patterns 1 to the recursive function F (step S201). Here, the input topology is set as a focused topology (step S202).

  The recursive function F generates a topology that occurs when one link is deleted in the noted topology (step S203). The recursive function F confirms whether the same topology exists among the generated topologies, and if so, counts the number of failure patterns of the topology (step S204). Here, when the recursive function F is focused on a topology in which a node having the number of output links of 0 exists (step S205, Yes), the number of combinations at that time is set to the topology of the initial state of the array A. The value is added to the value obtained by subtracting the number of links of the topology of interest from the number of links (step S206).

  Further, if the recursive function F does not pay attention to the topology in which the node having the output link number 0 exists (No in step S205), the recursive function F recurs the number obtained by multiplying the failure pattern number of the target topology by the count number. Input to the function F (step S207). If all the topologies are focused (step S208, Yes), the process is terminated. When not paying attention to all the topologies (step S208, No), pay attention to the next topology (step S209), and return to step S203 to repeat the processing.

[System configuration, etc.]
Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. Further, all or any part of each processing function performed in each device is realized by a CPU (Central Processing Unit) and a program analyzed and executed by the CPU, or hardware by wired logic. Can be realized as

  In addition, among the processes described in the present embodiment, all or part of the processes described as being automatically performed can be manually performed, or the processes described as being manually performed can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedure, control procedure, specific name, and information including various data and parameters shown in the above-described document and drawings can be arbitrarily changed unless otherwise specified.

[program]
As one embodiment, the link failure pattern number calculation system can be implemented by installing a link failure pattern number calculation program for executing the above link failure pattern number calculation as package software or online software in a desired computer. For example, by causing the information processing apparatus to execute the above link failure pattern number calculation program, the information processing apparatus can function as a link failure pattern number calculation system. The information processing apparatus referred to here includes a desktop or notebook personal computer. In addition, the information processing apparatus includes mobile communication terminals such as smartphones, mobile phones and PHS (Personal Handyphone System), and slate terminals such as PDA (Personal Digital Assistant).

  The link failure pattern number calculation system can also be implemented as a server device that uses a terminal device used by a user as a client and provides the client with the service related to the above link failure pattern number calculation. For example, the link failure pattern number calculation system is implemented as a server device that provides a link failure pattern number calculation service that receives the physical topology of the network and outputs the number of link failure patterns. In this case, the link failure pattern number calculation system may be implemented as a Web server, or may be implemented as a cloud that provides services related to the above link failure pattern number calculation by outsourcing.

  FIG. 18 is a diagram illustrating an example of a computer that realizes a link failure pattern number calculation system by executing a program. The computer 1000 includes a memory 1010 and a CPU 1020, for example. The computer 1000 also includes a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These units are connected by a bus 1080.

  The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM (Random Access Memory) 1012. The ROM 1011 stores a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. For example, a removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to a mouse 1110 and a keyboard 1120, for example. The video adapter 1060 is connected to the display 1130, for example.

  The hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. That is, a program that defines each process of the link failure pattern number calculation system is implemented as a program module 1093 in which a code executable by a computer is described. The program module 1093 is stored in the hard disk drive 1090, for example. For example, a program module 1093 for executing processing similar to the functional configuration in the link failure pattern number calculation system is stored in the hard disk drive 1090. The hard disk drive 1090 may be replaced by an SSD (Solid State Drive).

  The setting data used in the processing of the above-described embodiment is stored as program data 1094 in, for example, the memory 1010 or the hard disk drive 1090. Then, the CPU 1020 reads the program module 1093 and the program data 1094 stored in the memory 1010 and the hard disk drive 1090 to the RAM 1012 and executes them as necessary.

  The program module 1093 and the program data 1094 are not limited to being stored in the hard disk drive 1090, but may be stored in, for example, a removable storage medium and read out by the CPU 1020 via the disk drive 1100 or the like. Alternatively, the program module 1093 and the program data 1094 may be stored in another computer connected via a network (LAN (Local Area Network), WAN (Wide Area Network), etc.). The program module 1093 and the program data 1094 may be read by the CPU 1020 from another computer via the network interface 1070.

DESCRIPTION OF SYMBOLS 1 Link failure pattern number calculation system 10 Control part 11 Acquisition part 12 Deletion part 13 Pattern number addition part 14 Pattern update part 20 Storage part 21 Pattern storage part 22 Pattern number storage part

Claims (7)

In the topology of a network in which nodes are connected by links, a topology is generated when each link is deleted one by one, and in any of the generated topologies, when all nodes have links, A deletion unit that generates a topology when each link is deleted one by one, with a plurality of topologies having the same connection relation of links connected to the nodes being the same type of topology with respect to the topology in which all the nodes have links; ,
Of the topologies generated by the deletion unit, for the topology in which a node having no link exists, the number of link deletion patterns that can generate the topology is calculated for each number of links deleted until the generation of the topology. A pattern number adding unit to be summed;
A system for calculating the number of link fault patterns. For each topology, a pattern storage unit that stores the number of deletion patterns of links that can generate the topology;
Each time a topology is generated by the deletion unit, the number of topologies of the same type in which all the nodes have links and the topology before the links are deleted by the deletion unit stored in the pattern storage unit When the generated topology is stored in the pattern storage unit, the multiplied number is added to the number of deleted patterns of the generated topology, and the generated topology is multiplied. 2, further comprising: a pattern update unit that stores the multiplied number as the number of generated topology deletion patterns in the pattern storage unit when the pattern storage unit is not stored in the pattern storage unit. The link failure pattern number calculation system described. The pattern storage unit corresponds to the number of link deletion patterns that can generate the topology for each topology, and the number of link deletion patterns that can generate the topology before the topology is deleted by the deletion unit. Further memorize the failure pattern rate, which is the rate,
Each time the topology is generated by the deletion unit, the pattern update unit deletes the number of topologies of the same type in which all the nodes have links by the deletion unit stored in the pattern storage unit. When the number obtained by dividing the number of topological deletion patterns before being deleted is multiplied by the failure pattern ratio of the topology before the link is deleted by the deletion unit, and the generated topology is stored in the pattern storage unit The multiplied number is added to the failure pattern ratio of the generated topology, and when the generated topology is not stored in the pattern storage unit, the multiplied number is added to the failure pattern of the generated topology. The link failure pattern number calculation system according to claim 2, wherein the pattern storage unit stores the ratio as a ratio.   4. The delete unit according to claim 1, wherein the deletion unit deletes links in order from a topology having a large number of links deleted until generation of the topology among the input topologies. 5. The link failure pattern number calculation system described.   4. The delete unit according to claim 1, wherein the deletion unit deletes links in order from a topology having a smaller number of links deleted until generation of the topology among the input topologies. 5. The link failure pattern number calculation system described. In the topology of a network in which nodes are connected by links, a topology is generated when each link is deleted one by one, and in any of the generated topologies, when all nodes have links, Deletion step of generating a topology when each link is deleted one by one with a plurality of topologies having the same connection relation of links connected to the nodes as the same type of topology for all the nodes having links When,
Of the topologies generated by the deletion step, for the topologies where there are nodes that do not have links, the number of link deletion patterns that can generate the topology is calculated for each number of links deleted until the generation of the topology. Adding the number of patterns to be added;
A method for calculating the number of link fault patterns.   A link failure pattern number calculation program for causing a computer to function as the link failure pattern number calculation system according to any one of claims 1 to 5.

Images