Path provision assistance device, path provision assistance method, and program

Abstract

This path provision assistance device has: an initialization unit that, on the basis of data indicating, for each of a plurality of nodes of a physical network, the magnitude of the possibility of being selected as an end point of a logical transmission path set in the physical network, calculates an expected value that is selected as an end point of the transmission path for each set of two nodes among the plurality of nodes, and sets, for the portion of the sets having higher expected values, a secured number that is the number of transmission paths to be secured in advance; a selection unit that, on the basis of the expected value, probabilistically selects a plurality of first sets from among all of the sets; a route search unit that, for each of the first sets, searches for a route to which the first set can be connected under a constraint condition that sets having a relatively large number of hops, among the sets for which the secured number is set, are preferentially passed through; and a secured number adjustment unit that increases the secured number for sets for which a prescribed condition is satisfied by the number of times that the shortest routes of first sets for which a route satisfying the constraint condition was not found pass therethrough.

Description

Path provision support device, path provision support method, and program

[0001] The present invention relates to a path provision support device, a path provision support method, and a program.

[0002] The setup of a transmission path for a physical network is performed in response to an order from a user. Before the path can be provided, it requires time to design the route, input the settings into the equipment, and wait for the equipment's optical output level to stabilize.

[0003] It has been proposed to identify paths to be deployed in advance for immediate provision of paths according to orders by repeatedly generating a large number of orders according to the occurrence probability based on the population ratio of each region that could be the endpoint of the path, repeatedly issuing paths, and repeatedly adding missing paths and deleting excess paths (for example, Non-Patent Document 1).

[0004] Yoshifumi Kato, Masaru Miyoshi, "Proposal of a Resource Management Method for On-Demand Provision of Optical Paths," ICM Study Group, January 2023, January 19, 2023 [online], Internet <URL: https: / / www.ieice.org / publications / ken / summary.php?contribution_id=122732&society_cd=CS&ken_id=ICM&year=2023&present_date=2023-01-19&schedule_id=7569&lang=jp&expandable=1>

[0005] In order to be able to respond immediately to a variety of orders (i.e., to increase the number of paths that can be handled by combining pre-deployed paths), it is advantageous to pre-deploy a variety of short paths. As a result, the method of Non-Patent Document 1 results in a large number of short paths being deployed. When providing paths according to orders by combining pre-deployed short paths, the number of connections between pre-deployed paths increases between paths according to orders, which requires a large number of facilities and devices for connecting the paths, resulting in an increase in the provider's facility construction costs.

[0006] The present invention has been made in view of the above points, and has an object to support the improvement of the efficiency of providing network transmission paths.

[0007] Therefore, in order to solve the above problem, the path provision support device has an initialization unit configured to calculate an expected value for each pair of two nodes among the plurality of nodes to be selected as an endpoint of the transmission path set in the physical network based on data indicating, for each of the plurality of nodes in the physical network, the likelihood that the node will be selected as an endpoint of the transmission path, and to set a reservation number, which is the number of transmission paths to be reserved in advance, for some of the pairs with the highest expected values; a selection unit configured to probabilistically select a plurality of first pairs from all of the pairs based on the expected value; a route search unit configured, for each of the first pairs, to search for a route that can connect the first pair under the constraint that, among the pairs for which the reservation number is set, the pair with a relatively large number of hops is preferentially passed through; and a reservation number adjustment unit configured to increase the reservation number for the pair for which no route satisfying the constraint is found, whereby the number of times the shortest route passes through the first pair satisfies a predetermined condition, and the path provision support device selects a plurality of first pairs probabilistically based on the expected value.

[0008] It can help improve the efficiency of providing network transmission paths.

[0009] 1 is a diagram illustrating an example of a hardware configuration of a path provision support device 10 according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an example of a functional configuration of the path provision support device 10 according to an embodiment of the present invention. FIG. 3 is a flowchart illustrating an example of a processing procedure executed by the path provision support device 10. FIG. 4 is a diagram illustrating a first example of network information. FIG. 5 is a diagram illustrating a second example of network information. FIG. 6 is a diagram illustrating an example of the configuration of demand forecast source data. FIG. 7 is a diagram illustrating the relationship between edges and paths. FIG. 8 is a diagram illustrating simulation parameters. FIG. 9 is a diagram illustrating an example of the configuration of a recorded data storage unit 124. FIG. 10 is a diagram illustrating the inventory quantity of paths. FIG. 11 is a diagram illustrating a node pair in which 1 is added to the shortage number in the case of two hops. FIG. 12 is a diagram illustrating a node pair in which 1 is added to the shortage number in the case of three hops. FIG. 13 is a diagram illustrating a node pair in which 1 is added to the shortage number in the case of four hops. FIG. 14 is a diagram illustrating a node pair in which the shortage number is recorded according to multiple orders when there is no constraint according to the number of hops of the order path. FIG. 15 is a diagram illustrating a node pair in which the shortage number is recorded according to multiple orders when there is a constraint according to the number of hops of the order path.

[0010] An embodiment of the present invention will now be described with reference to the accompanying drawings. Fig. 1 is a diagram showing an example of the hardware configuration of a path provision support device 10 according to an embodiment of the present invention. The path provision support device 10 in Fig. 1 includes a drive device 100, an auxiliary storage device 102, a memory device 103, a processor 104, and an interface device 105, all of which are interconnected via a bus B.

[0011] The program that realizes the processing in the path provision support device 10 is provided by a recording medium 101 such as a CD-ROM. When the recording medium 101 storing the program is set in the drive device 100, the program is installed from the recording medium 101 to the auxiliary storage device 102 via the drive device 100. However, the program does not necessarily have to be installed from the recording medium 101, but may be downloaded from another computer via a network. The auxiliary storage device 102 stores the installed program as well as necessary files, data, etc.

[0012] When an instruction to start a program is received, the memory device 103 reads and stores the program from the auxiliary storage device 102. The processor 104 is a CPU or a GPU (Graphics Processing Unit), or a CPU and a GPU, and executes functions related to the path provision support device 10 in accordance with the program stored in the memory device 103. The interface device 105 is used as an interface for connecting to a network.

[0013] FIG. 2 is a diagram showing an example of the functional configuration of a path provision support device 10 according to an embodiment of the present invention. In FIG. 2, the path provision support device 10 includes an initialization unit 11, an order generation unit 12, a route search unit 13, a reservation number adjustment unit 14, and an output unit 15. Each of these units is implemented by a processor 104 executing one or more programs installed in the path provision support device 10. The path provision support device 10 also utilizes a network information storage unit 121, a parameter storage unit 122, a demand forecast source data storage unit 123, and a recorded data storage unit 124. Each of these storage units can be implemented using, for example, an auxiliary storage device 102 or a storage device connectable to the path provision support device 10 via a network.

[0014] 2, the path provision support device 10 generates path orders stochastically and executes a simulation to predict what transmission paths (hereinafter simply referred to as "paths") should be preset for the physical network that is the target of the path orders (hereinafter referred to as the "target network") to enable quick response to the actual path orders. In other words, the path provision support device 10 selects paths that, if preset, can quickly respond to the actual path orders.

[0015] The following describes the processing procedure executed by the path provision support device 10. FIG.

[0016] In step S101 , the initialization unit 11 acquires network information of the target network from the network information storage unit 121 .

[0017] FIG. 4 is a diagram showing a first example of network information. The network information is information indicating the topology of the target network, etc. In FIG. 4, the network information is expressed using a graph consisting of a plurality of nodes and edges connecting the nodes. In the graph, a node is, for example, a city or region where a data center or the like is located. An edge is a physical line connecting nodes. A numerical value assigned to an edge is the distance of the edge (the distance between the nodes connected by the edge).

[0018] The network information may be expressed in a structure other than a graph structure, as long as the format is capable of expressing the content shown in FIG.

[0019] Fig. 5 is a diagram showing a second example of network information. In Fig. 5, the network information is expressed in a matrix format. An element (component) having a value indicates that the node on the row and the node on the column of the element are connected by an edge having that value as a distance.

[0020] Next, the order generating unit 12 acquires the demand forecast source data from the demand forecast source data storage unit 123 (S102).

[0021] FIG. 6 is a diagram showing an example of the configuration of demand forecast source data. Demand forecast source data refers to data used to predict the likelihood that each node will be selected as a path endpoint in a path order (the level of path order demand for each node). FIG. 6 shows the population of the region to which each node belongs as an example of demand forecast source data. In this embodiment, the population of each node is used as a source of prediction for the probability that each node will be selected as a path endpoint. This is because the more populated a location is, the more likely it is to be designated as a path endpoint. In other words, it is considered that a path is more likely to be established between cities with relatively large populations.

[0022] A path is a logical communication route that connects two nodes via one or more edges (physical lines). Fig. 7 is a diagram for explaining the relationship between edges and paths.

[0023] In Figure 7, edges are represented by cylinders, and paths are represented by solid lines. Two paths are shown in Figure 7. One is a path connecting node A and node D, and its route is ABCD. The other is a path connecting node A and node F, and its route is ABF.

[0024] To establish a communication route using a path, a network user issues an order specifying the two end points (two nodes) of the path.

[0025] Next, the initialization unit 11 acquires simulation parameters from the parameter storage unit 122 (S103).

[0026] 8 is a diagram showing simulation parameters. The simulation parameters are parameters for controlling thresholds and the like used in subsequent processes. Details of each parameter will be explained when each parameter is used.

[0027] Next, the initialization unit 11 calculates, based on the network information (FIG. 4 or FIG. 5), for all possible pairs of two nodes that can be connected in the target network (hereinafter, pairs of two nodes are referred to as "node pairs"), the expected value of an order for a path connecting the node pair (the probability of occurrence of the order) based on the demand forecast source data (FIG. 6) (S104). Specifically, the initialization unit 11 calculates the total population of the nodes related to each node pair based on the demand forecast source data (FIG. 6), and calculates the expected value (probability of occurrence) of an order for each node pair so that it is proportional to the ratio of the total population of each node pair. Note that all possible node pairs that can be connected in the target network are defined as follows, assuming that each node can be connected to all other nodes and the number of nodes in the target network is n: ｎ C ２There are 1000 possible node pairs. That is, in this embodiment, the start point and end point of the path (the order of the nodes in the node pair) are not distinguished. The pair of node A and node B and the pair of node B and node A are treated as the same node pair. Note that all possible node pairs in the target network are a set of node pairs that may be designated as both end points of the path in the path order.

[0028] Next, the initialization unit 11 sorts the node pairs in descending order of expected value (S105).

[0029] Next, the initialization unit 11 sets the secured number to 1 for the top N node pairs (i.e., some of the top pairs) in descending order of expected value, and sets the secured number to 0 for the N+1th node pairs and below (S106). The secured number is a parameter constituting data recorded and stored for each node pair (hereinafter referred to as "recorded data"), and is the number of transmission paths that should be secured in advance to directly connect the two nodes related to the node pair via the shortest route. The secured number is set as the initial value of the path inventory number updated for the node pair related to the path used in the trial. The path inventory number for a certain node pair refers to the number of times that the node pair can be used as a path according to the path order (hereinafter referred to as "order path"). In other words, the path inventory number for a certain node pair refers to the number of paths virtually set between the node pair. The secured number for each node pair is recorded in the recorded data storage unit 124.

[0030] FIG. 9 is a diagram showing an example of the configuration of the recorded data storage unit 124. In FIG. 9, the recorded data storage unit 124 stores the number of reserved nodes, the number in stock, the number in use, the number of missing nodes, and the lock for each of all possible node pairs. The recorded data storage unit 124 also stores the route of each path for each node pair for which the number of reserved paths is one or more. The route of a path is expressed by the sequence of nodes through which the path passes. Note that in this embodiment, the route of a single path is the shortest route between the two nodes connected by the path.

[0031] In step S106, the value of the number of allocations for each node pair is recorded in the record data storage unit 124.

[0032] 10 is a diagram for explaining the number of paths in stock. In FIG. 10, it is shown that the number of paths between node pair A and D is three, and the number of paths between node pair A and F is one. In this case, the number of paths in stock between node pair A and D is three, and the number of paths in stock between node pair A and F is one.

[0033] Next, the initialization unit 11 locks the node pair whose reserved number is set to 1 (S107). A lock is information for preventing the reserved number of the node pair from being set to 0 in the process described below. A certain node pair is locked, for example, by recording "1" in the lock column for the node pair in the record data storage unit 124 (FIG. 9).

[0034] Next, the initialization unit 11 initializes the number of uses and the number of shortages in the recorded data storage unit 124 (FIG. 9) for each node pair to 0 (S108). The number of uses and the number of shortages are also recorded for each node pair, along with the number of stocks, and are parameters that constitute the recorded data updated during the simulation. The number of uses for a certain node pair refers to the number of times that node pair is used for the path order in which a path between the node pair occurs in the simulation (the number of times a path related to the path order passes through the node pair). The number of shortages for a certain node pair refers to the number of times an order path could not be connected between the node pair because there was no path in stock (because the path stock number = 0). S108 Next, the initialization unit 11 assigns the number of reserved nodes for the node pair to the stock numbers for all node pairs (S109).

[0035] Next, the order generation unit 12 generates one path order probabilistically based on the demand forecast source data (FIG. 6) (S110). Probabilistic generation of one path order is synonymous with probabilistic selection of one node pair. Probabilistic selection of one node pair means probabilistically selecting two mutually distinct nodes from multiple nodes in the target network. When selecting the two nodes, the expected value calculated in step S104 based on the demand forecast source data (FIG. 6) is used. In other words, the order generation unit 12 selects two nodes (node ​​pair) with a probability based on the expected value. Hereinafter, the node pair related to the generated path order will be referred to as an "order pair."

[0036] Next, the route search unit 13 searches for the shortest route between the order pairs without imposing a constraint that only the path inventory is used (S111). The number of hops (number of edges) of the shortest route between the order pairs is hereinafter referred to as the "number of hops of the order path."

[0037] Next, the route search unit 13 searches for a route that can connect the order pairs using only the path inventory (S112). For example, the route search unit 13 searches for a route between the order pairs under a constraint that the route only passes between two nodes where the path inventory is 1 or more. At this time, the route search unit 13 changes the constraint as follows depending on the number of hops of the order path, in order to restrict the route search to preferentially passing through paths with a relatively large number of hops.

[0038] When the number of hops of the order path is 1 (very short distance), the route search unit 13 sets a constraint that the route must pass through a path (node ​​pair) with a hop count of 1 and an inventory quantity of 1 or more.

[0039] When the number of hops of the order path is two or more and three or less (short distance), the route search unit 13 adds as a constraint that the order path must be connectable (configurable) with two or less node pairs (one path or two paths) with an inventory of one or more. Because the constraint is that the order path must be connectable with two or less node pairs, it is acceptable for the route to be longer than a route that can be reached with three paths (a detour route). Note that the number of hops of a certain node pair refers to the number of hops of the shortest route of the node pair.

[0040] When the number of hops of the order path is four or more (long distance), the route search unit 13 adds a constraint that the order path must be connectable (configurable) with two or less node pairs (one path or two paths) with two or more hops and an inventory of one or more. Therefore, a route that does not pass through a node pair with one hop is searched for. However, even in this case, a detour route search is permitted.

[0041] If the search for the route (order path) is successful (Yes in S113), the route search unit 13 updates the usage count and inventory count of the node pairs related to each path used in the order path (S114). Specifically, the route search unit 13 adds 1 to the usage count of the node pair and subtracts 1 from the inventory count of the node pair. Therefore, if the path of the searched route is one hop, the usage count and inventory count of the node pairs that are both endpoints of the one hop are updated. If the path of the searched route is two hops, the usage count and inventory count of the two node pairs that make up the two hops are updated.

[0042] On the other hand, if the search for the route (order path) fails (No in S113), the route search unit 13 adds 1 to the number of shortages of the following node pairs among the node pairs that make up the shortest route between the order pairs (hereinafter referred to as the "stock shortage route") according to the number of hops of the order path (S115). In other words, 1 is added to the number of shortages of the node pairs that caused the failure to connect the order path.

[0043] If the number of hops in the order path is 1, the node pair is 1, so 1 is added to the missing number of the node pair.

[0044] If the number of hops in the order path is 2, 1 is added to the missing number of all node pairs that can form the shortest path between the order pairs.

[0045] 11 is a diagram for explaining node pairs for which 1 is added to the shortage number in the case of two hops. In FIG. 11, the node pairs for which 1 is added to the shortage number when the search for the two-hop shortest path of ABC fails are shown as AB, BC, and AC.

[0046] If the number of hops in the order path is 3, 1 is added to the missing number of all node pairs that can connect the shortest path between the order pairs with two or fewer paths (i.e., node pairs that can form the shortest path with two or fewer paths).

[0047] 12 is a diagram for explaining node pairs for which 1 is added to the shortage number in the case of three hops. In FIG. 12, the node pairs for which 1 is added to the shortage number when the search for the three-hop shortest path of ABCD fails are shown as A-C, CD, AB, BD, and A-D.

[0048] If the number of hops between the node pair is four or more, one is added to the missing number of all node pairs that can connect the shortest path between the order pairs with two or more hops but no more than two paths (i.e., node pairs with two or more hops that can form the shortest path with two or less paths).

[0049] 13 is a diagram for explaining node pairs for which 1 is added to the shortage number in the case of four hops. In FIG. 12, the node pairs for which 1 is added to the shortage number when the search for the four-hop shortest path of ABCDF fails are shown as AD, DF, AC, CF, and AF.

[0050] Following step S114 or S115, the order generation unit 12 determines whether a specified number of orders have been generated (S116). The specified number of orders refers to the "number of orders per trial" in the simulation parameters (FIG. 8), which is 50 in this embodiment. That is, in this embodiment, 50 orders are processed in one trial. Therefore, in this case, the order generation unit 12 determines whether step S110 and subsequent steps have been repeated 50 times. Note that the number of orders that have been generated is initialized to 0 at the start of the trial (at step S109).

[0051] If the specified number of orders has not been generated (No in S116), steps S110 and subsequent steps are repeated. If the specified number of orders has been generated (No in S116), the reservation quantity adjustment unit 14 determines whether the number of trials (step S109 and subsequent steps) has reached the "number of trials" in the simulation parameters ( FIG. 8 ) (S117). According to FIG. 8, the "number of trials" in this embodiment is 10. Therefore, it is determined whether steps S109 and subsequent steps have been repeated 10 times. The number of trials is initialized to 0 before the start of the trial (at the timing of step S108). If the number of trials is less than the "number of trials" (No in S117), steps S109 and subsequent steps are repeated. At this time, the inventory quantity is initialized to the reservation quantity in step S109, but the usage quantity and shortage quantity are not initialized. In other words, the usage quantity and shortage quantity are cumulative values ​​for the 10 trials. Therefore, the number of uses of a certain node pair is the number of order paths that use the inventory of the path of that node pair among the order paths searched in 10 trials, while the number of shortages of a certain node pair is a value that indicates the shortage of the number of reserved nodes for that node pair relative to the number of inventory shortage routes that pass through that node pair among the inventory shortage routes searched in 10 trials.

[0052] If the number of attempts is less than the "number of attempts" (No in S117), the attempts from step S109 onwards are repeated. If the number of attempts has reached the "number of attempts" (Yes in S117), the allocation number adjustment unit 14 executes loop processing L1 including steps S118 to S124 for each node pair. The node pair being processed in loop processing L1 is hereinafter referred to as the "target pair".

[0053] In step S118, the reservation number adjustment unit 14 determines whether the reservation number of the target pair is 1 or greater. If the reservation number of the target pair is 1 or greater (Yes in S118), the reservation number adjustment unit 14 determines whether the utilization rate of the target pair is equal to or less than the path removal determination threshold (S119). The path removal determination threshold is a threshold that is set in advance as a simulation parameter ( FIG. 8 ), and is 0.2 in this embodiment. The utilization rate of the target pair is calculated using the following formula: Utilization rate of target pair = Number of used target pairs / (Number of reserved target pairs × Number of attempts) For example, suppose the number of used target pairs is 1 and the number of reserved target pairs is 6. In this embodiment, the number of attempts is 10, so the utilization rate of the target pair in this case is as follows: Utilization rate of the target pair = 1 / (6 × 10) = 0.0166... ​​If the utilization rate of the target pair is equal to or less than the path reduction determination threshold (if the number of order paths passing through the target pair is small enough to satisfy a predetermined condition with respect to the reserved number of the target pair) (Yes in S119), the reserved number adjustment unit 14 determines whether the target pair is locked by referring to the recorded data storage unit 124 ( FIG. 9 ) (S120). That is, it is determined whether the target pair is a node pair whose reserved number was set to 1 in step S106. In the recorded data storage unit 124 ( FIG. 9 ), if the lock value for the target pair is 1, the target pair is locked, and if the lock k value for the target pair is 0, the target pair is not locked.

[0054] If the target pair is not locked (No in S120), the process proceeds to step S122. If the target pair is locked (Yes in S120), the allocation number adjustment unit 14 determines whether the allocation number of the target pair is 2 or more (S121).

[0055] If the secured number of the target pair is two or more (Yes in S121), or if the answer is No in step S120, the secured number adjustment unit 14 reduces the secured number of the target pair (S122). In this embodiment, the secured number of the target pair is reduced by one. This is because there is a possibility that the secured number of the target pair is surplus. In this case, steps S123 and S124 are not executed for the target pair. Note that if the target pair is locked, the secured number of the target pair is reduced only if the secured number of the target pair is two or more, which makes it possible to prevent the secured number of the target pair from becoming zero.

[0056] On the other hand, if the number of reserved paths for the target pair is less than 1 (No in S118), or if the usage rate of the target pair exceeds the path reduction determination threshold (No in S119), the reserved-path-number adjustment unit 14 determines whether the number of paths insufficient for the target pair exceeds the path addition determination threshold (S123). The path addition determination threshold is a threshold that is set in advance as a simulation parameter ( FIG. 8 ), and is 5 in this embodiment. If the number of paths insufficient for the target pair is equal to or less than the path addition determination threshold (No in S123), step S124 is not executed for the target pair.

[0057] If the number of shortages in the target pair exceeds the path addition determination threshold (i.e., if the inventory shortage route passing through the target pair is greater than the number of reserved paths in the target pair by more than the path addition determination threshold) (Yes in S123), the reserved number adjustment unit 14 increases the number of reserved paths in the target pair by a maximum of m (upper limit s) within the edge capacity limit (S124). Here, the value of m is as follows: m = number of shortages in the target pair / path addition determination threshold Furthermore, the value of m is the "upper limit number of reserved paths" that is set in advance as a simulation parameter (FIG. 8), and is 3 in this embodiment.

[0058] Furthermore, "within the capacity limit of an edge" means that the "maximum number of paths that can be passed per edge" that is preset as a simulation parameter (FIG. 8) for all edges belonging to the route of the target pair does not exceed the "maximum number of paths that can be passed per edge." According to FIG. 8, in this embodiment, the "maximum number of paths that can be passed per edge" is 20. For example, in FIG. 10, if the target pair is A-D, the remaining capacity for B-C and C-D is 17, but the remaining capacity for A-B is 16, so 16 is a value within the capacity limit of the edge. Note that the remaining capacity of a certain edge is obtained by minus the "maximum number of paths that can be passed per edge" - the number of reserved edges.

[0059] For example, if the expansion determination threshold is 5 and the shortage number is 24, then m = 24 (shortage number) / 5 (path expansion determination threshold) = 4 (decimals are rounded down). However, if the "maximum number of paths to be expanded" is 3, the increase in the number of reserved paths will be 3. Furthermore, if the remaining capacity of any edge belonging to the route of the target pair is 2, the increase in the number of reserved paths will be 2.

[0060] It is also possible to set m = (number of shortages - path addition determination threshold), but in this case, there is a possibility that the number of reserved paths will increase suddenly. In this embodiment, the number of reserved paths is gradually increased by repeating trials, so the increase in the number is made gradual by division.

[0061] In this way, in the loop process L1, the number of reserved nodes is decreased for node pairs with a surplus of paths, and the number of reserved nodes is increased for node pairs with a shortage of paths.

[0062] When the loop process L1 has been executed for all node pairs, the reservation number adjustment unit 14 determines whether the number of times the inventory review process has been executed (repetition count) has reached a specified number of reviews (S125). Here, repetition of the loop process L1 for all node pairs corresponds to one execution of the inventory review process. Therefore, when step S125 is executed for the first time, the number of times the inventory review process is executed is 1. The specified number of reviews is the "number of path inventory reviews" that is preset as a simulation parameter (FIG. 8), and is 20 in this embodiment.

[0063] If the number of times the inventory review process has been executed is less than the specified number of reviews (No in S125), steps S108 and subsequent steps are repeated. Therefore, the usage count and shortage count for all node pairs are initialized to 0 (S108), the inventory count for all node pairs is initialized to the reserved count (S109), and steps S110 and subsequent steps are repeated. That is, in this embodiment, 10 trials including 50 order generation are repeated, and the path inventory review process for each of the 10 trials is repeated 20 times. Therefore, order generation is executed 50 x 10 x 20 = 10,000 times, and trials are executed 10 x 20 = 200 times.

[0064] Since the reserved quantity may change with each inventory review process, the reserved quantity for each node pair may differ from the previous one in the subsequent step S109. In other words, the reserved quantity adjusted based on the previous trial is used as the initial value of the inventory quantity for the next 10 trials.

[0065] If the number of executions of the inventory review process is equal to or greater than the predetermined number of reviews (Yes in S125), the output unit 15 outputs the following as a result of the simulation: ｎ C ２ For each node pair, the route and the number of reserved paths are output (S126).

[0066] A user who obtains such output can pre-configure paths (pre-order configuration) for the target network based on the output. The route of the path is the route (shortest route) output for the node pair. The number of reserved paths for each node pair is the number of paths that need to be reserved in advance between each node pair based on the order demand forecast. Therefore, the number of reserved paths is considered to indicate the likelihood of an actual order (or the predicted number of orders). Therefore, the user may pre-configure paths for each node pair in the target network in accordance with the number of reserved paths for that node pair. Alternatively, the user may treat the number of reserved paths for each node pair as a ratio and pre-configure a number of paths between each node pair according to the ratio of the number of reserved paths. In this case, the user may pre-configure only node pairs whose number of reserved paths is equal to or greater than a threshold. Such pre-configuration allows for quick response to two nodes specified in the actual path order. This is because a path between the two nodes may already be configured. Furthermore, even if a path between the two nodes has not been configured, a path between the two nodes can be provided by connecting already configured paths. Even in this case, the path can be provided more quickly than when the path is set from scratch after receiving a path order.

[0067] Furthermore, according to this embodiment, in step S112, when the number of hops of the order path is two or more, a route that can be reached with two or fewer paths is searched for, and when the number of hops of the order path is four or more, a route using a path with one hop is not searched for, so that it is possible to prevent a decrease in the number of hops of a path that would otherwise have a large reservation count (i.e., a path that can be pre-set). The reservation count is determined based on the shortage count. Hereinafter, a difference in the shortage count when a path order occurs will be described between the case where, in the route search in step S112, constraints based on the number of hops of the order path (such as a requirement that the route can be reached with two or fewer paths or not searching for routes using a path with one hop) are eliminated (hereinafter referred to as "a case where there is no constraint based on the number of hops of the order path") and this embodiment (hereinafter referred to as "a case where there is a constraint based on the number of hops of the order path").

[0068] 14 is a diagram illustrating node pairs for which shortages are recorded according to multiple orders when there are no constraints according to the number of hops in the order path. In FIG. 14, for path order 1 for node pair A-D with a 3-hop count, path order 2 for node pair A-E with a 4-hop count, and path order 3 for node pair A-G with a 6-hop count, the node pairs (paths) for which shortages are recorded when the route search fails in step S112 are assigned the symbols d1, d2, and d3. Hereinafter, a line with the symbol dx (x = 1, 2, 3) attached thereto will be referred to as a "line sx."

[0069] The node pair indicated by line s3 is a node pair whose deficit number becomes 3 with a path order of 3. In FIG. 14, the longest hop count of line s3 is 3 for node pair A-D, whose deficit number is recorded with path order 3.

[0070] The node pair indicated by line s2 is a node pair whose deficit becomes 2 with three path orders. In FIG. 14, the longest hop count of line s2 is 4 for node pair A-E, whose deficit is recorded with path orders 2 and 3.

[0071] The node pair indicated by line s1 is a node pair whose deficit becomes 1 with a path order of 3. In FIG. 14, the longest hop count of line s1 is 6 for node pair A-G, whose deficit is recorded with a path order of 3.

[0072] As is clear from the above, if there is no restriction according to the number of hops in the order path, the shortage will be greater for node pairs with a shorter number of hops.

[0073] Even in cases where it is appropriate to use a long-distance path (a path with a large number of hops), the shortage is added to the node pairs with a short number of hops, resulting in an increase in the shortage of node pairs with a short number of hops.On the other hand, the shortage of long-distance paths is small.

[0074] Since the number of reserved nodes for node pairs with a large shortage is increased, the number of paths allocated increases, mainly for node pairs with short hop counts, and the number of paths allocated decreases for medium- and long-hop paths. In addition, the increase in paths for node pairs with short hop counts may cause resource depletion, which may hinder the generation of medium- and long-hop paths.

[0075] 15 is a diagram for explaining node pairs for which shortage numbers are recorded according to multiple orders when there are constraints according to the number of hops in the order path. In Fig. 15, the meanings of path orders 1 to 3 and lines s1, s2, and s3 are the same as in Fig. 14. However, in Fig. 15, node pairs for which shortage numbers are not recorded are indicated by dashed lines compared to Fig. 14.

[0076] In this embodiment, in path order 1 for a short distance path of three hops or less, the shortage number is recorded for a node pair of a short distance of one hop or two hops.

[0077] On the other hand, in path orders 2 and 3, which are four hops or more, the number of records of the shortage for node pairs with short distances of one or two hops decreases.

[0078] This creates an opportunity to add paths with an appropriate number of hops according to the path order distance, such that the shortage of short-hop node pairs increases when a short-hop path order is used, and the shortage of medium- to long-hop node pairs increases when a long-hop path order is used. As a result, it is possible to equalize the opportunities for generating short-distance paths and medium- to long-distance paths, preventing resource depletion due to an increase in short-distance paths and enabling the increase of medium- to long-distance paths.

[0079] As described above, according to this embodiment, it is possible to support the improvement of the efficiency of providing transmission paths in a network. For example, it is possible to support the provision of transmission paths that are fast and capable of suppressing increases in facility construction costs in response to any path order of a user. Furthermore, it is possible to limit the amount of available resources in accordance with the requirements of the network (the upper limit of the number of paths that can be passed per edge), thereby preventing excessive resource allocation.

[0080] By providing paths quickly, it becomes possible to utilize temporary alternative paths when a transmission path fails. For example, it is possible to search for a combination of paths that can be immediately provided between two points when a failure occurs, provide that combination of paths, and then switch back to the original path after the failure is restored.

[0081] It can also be used to increase the speed of paths. For example, if the bandwidth of an existing path is insufficient, a combination of paths with a larger bandwidth can be searched for and switched to, allowing for an immediate transition to an increased speed path.

[0082] In this embodiment, an example has been shown in which a path that is likely to be used is selected by the Monte Carlo method, but other methods may be adopted as the path selection algorithm.

[0083] In this embodiment, the order generating unit 12 is an example of a selecting unit.

[0084] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims.

[0085] 10 Path provision support device 11 Initialization unit 12 Order generation unit 13 Route search unit 14 Allocation number adjustment unit 15 Output unit 100 Drive device 101 Recording medium 102 Auxiliary storage device 103 Memory device 104 Processor 105 Interface device 121 Network information storage unit 122 Parameter storage unit 123 Demand forecast source data storage unit 124 Recorded data storage unit B Bus

Claims

1. An initialization unit configured to calculate an expected value of being selected as an endpoint of a logical transmission path set in a physical network for each pair of two nodes among the plurality of nodes based on data indicating the likelihood of being selected as an endpoint of the logical transmission path for each of the plurality of nodes in the physical network, and to set a reservation number, which is the number of transmission paths to be reserved in advance, for a part of the pairs with higher expected values; a selection unit configured to probabilistically select a plurality of first pairs from among all the pairs based on the expected value; a path search unit configured to search for a path capable of connecting each of the first pairs, with the constraint that, for each of the first pairs, a pair with a relatively large number of hops among the pairs for which the reservation number is set is preferentially passed through; and a reservation number adjustment unit configured to increase the reservation number of a pair for which the shortest path of the first pair for which no path satisfying the constraint is found passes through a number of times that satisfies a predetermined condition. A path provision support device characterized by comprising these components.

2. The path provision support device according to claim 1, wherein the path search unit is configured to add, to the constraint condition, that the first pair can be connected by two or fewer of the pairs, and search for a path when the number of hops of the shortest path of the first pair is 2 or more.

3. The path provision support device according to claim 2, wherein the path search unit is configured to add, to the constraint condition, that the pair with a hop count of 1 is not passed through, and search for a path capable of connecting the first pair when the number of hops of the shortest path of the first pair is 4 or more.

4. The path provision support device according to claim 1, wherein the reservation number adjustment unit is configured to increase the reservation number of a pair that can be configured with two or fewer of the shortest paths when the first pair with a hop count of 2 or more in the shortest path is not found.

5. The path provision support device according to claim 1, wherein the reservation number adjustment unit is configured to increase the reservation number of a pair that can be configured with two or fewer of the shortest paths and has a hop count of 2 or more when the first pair with a hop count of 4 or more in the shortest path is not found.

6. The guarantee number adjustment unit is configured to decrease the guarantee number of the group in which the number of times the path passes is small enough to satisfy a predetermined condition with respect to the guarantee number, and when the expected value of the group is part of the upper part, not to decrease the guarantee number to zero. The path providing support device according to any one of claims 1 to 5, characterized in that.

7. Based on data indicating the likelihood of being selected as an endpoint of a logical transmission path set in a physical network for each of a plurality of nodes in the physical network, an expected value of being selected as an endpoint of the transmission path is calculated for each pair of two nodes among the plurality of nodes, and for the group in which the expected value is part of the upper part, an initialization procedure for setting a guarantee number, which is the number of transmission paths to be secured in advance; a selection procedure for probabilistically selecting a plurality of first groups based on the expected value from among all the groups; a path search procedure for searching for a path that can connect the first group, with the constraint condition that the group with a relatively large number of hops among the groups for which the guarantee number is set is preferentially passed for each first group; and a guarantee number adjustment procedure for increasing the guarantee number of the group in which the number of times the shortest path of the first group that does not satisfy the constraint condition passes satisfies a predetermined condition. A path providing support method, characterized in that a computer executes the above.

8. Based on data indicating the likelihood of being selected as an endpoint of a logical transmission path set in a physical network for each of a plurality of nodes in the physical network, an expected value of being selected as an endpoint of the transmission path is calculated for each pair of two nodes among the plurality of nodes, and for the group in which the expected value is part of the upper part, an initialization procedure for setting a guarantee number, which is the number of transmission paths to be secured in advance; a selection procedure for probabilistically selecting a plurality of first groups based on the expected value from among all the groups; a path search procedure for searching for a path that can connect the first group, with the constraint condition that the group with a relatively large number of hops among the groups for which the guarantee number is set is preferentially passed for each first group; and a guarantee number adjustment procedure for increasing the guarantee number of the group in which the number of times the shortest path of the first group that does not satisfy the constraint condition passes satisfies a predetermined condition. A program, characterized in that the computer is caused to execute the above.

Images