JP4724219B2 - Shortest path calculation device, shortest path calculation method, and program - Google Patents

Description

  The present invention relates to a method for calculating a route to a destination, and more particularly, to a shortest route calculation device that calculates a shortest route to a destination, a shortest route calculation method, and a program for causing a computer to execute the method.

  Changes in the way of linking nodes and bases in the network and the network link laying status (topology) may result in passage of points that could not be passed until then, or vice versa. There is a case. When it is necessary to always maintain the shortest route from the starting point to the destination, it is necessary to calculate the shortest route in the changed topology as the path topology changes. The conventional shortest path calculation device recalculates all the shortest paths connecting two bases by using topology information.

  Non-patent document 1 describes the calculation method. This method is known as Dijkstra's algorithm.

  It should be noted that there is a two-node pair in which there is no difference in the shortest path before and after the topology change. In addition, if there is a shortest path connecting between different bases that is calculated after the topology change, there are many cases in which there is no difference between this shortest path and the shortest path between bases to be calculated.

  However, in the conventional shortest path calculation device, since the calculation of the shortest path is performed using only information on the laying status, it is impossible to use the existing shortest path information if there is no difference. is there. For this reason, it is necessary to recalculate all the shortest paths, resulting in a problem that the calculation time becomes long.

  Usually, in such a route calculation method, a number called a cost is assigned to a transmission path or a path connecting between bases, and a route that minimizes the cost sum of links included in the route that has passed is calculated. Non-Patent Document 2 describes a method of calculating using the shortest path information before a topology change when the cost of a link changes.

By using this method, it is possible to solve the problem that the calculation time is increased by recalculating all the shortest paths described above.
E. W. Dijkstra: A note on two problems in connection with graphs. In Numerisch Mathematic 1, (1959), S.N. 269-271. B. Xiao, et al. “Dynamic update of shorttest path tree in OSPF,” IEEE Parallel Architectures Algorithms and Network, pp. 18-23, May 2004.

  In Non-Patent Document 2, calculation is performed only when the cost of a link existing in the topology changes.

  In addition to this, a link connecting nodes may be added to the change in topology, or a link and a base may be added simultaneously. In the technique disclosed in Non-Patent Document 2, a topology change due to the addition of a transmission line that did not exist before the topology change is not handled as a trigger for starting calculation, and recalculation is not performed. Therefore, it is necessary to add a procedure to calculate the shortest path in this case.

  The present invention has been made to solve the above-described problems of the technology, and is the shortest possible recalculation of the shortest route when a node or link is added or deleted from the network. An object of the present invention is to provide a route calculation device, a shortest route calculation method, and a program for causing a computer to execute the method.

The shortest path calculation apparatus of the present invention for achieving the above object is a shortest path calculation apparatus for calculating the shortest path between nodes in a network,
A storage unit for storing a calculation target tree including information on a node that can be reached from a starting point node that is a starting point node by following an upstream / downstream relationship between the nodes, and an upstream / downstream connection relationship between the nodes;
When there is a change in the topology of the network, the calculation target tree is set. When the change in the topology is addition of a new first link, the first link is a node in the calculation target tree. For the first and second nodes connected in the above, the first node via the path from the origin node to the second node along the calculation object tree and the path of the first link The distance between a first route that is a route that reaches the first node and a second route that is reached from the origin node along the calculation target tree to the first node is compared, and the distance from the second route is more than the second route. If the distance of the first route is short, the first route is determined as a candidate for the shortest route from the origin node to the first node, and the deletion is performed when the change in topology is deletion of a link. When the information on the downstream side of the calculated link is deleted from the calculation target tree, and the second link connecting the third node in the calculation target tree and the fourth node not belonging to the calculation target tree is found, A route consisting of the second link and a route reaching the third node along the calculation target tree from the origin node is taken as a candidate of the shortest route from the origin node to the fourth node, and A processor for identifying a candidate having the shortest distance from a plurality of candidates and determining the identified candidate as the shortest path if there are a plurality of candidates for the shortest path with respect to the route to the first or fourth node;
I have a,
The processor is
If there is information corresponding to an upstream / downstream relationship between nodes corresponding to a downstream path that is opposite to the origin node for the relay node that is the first or the fourth node, the information Rewrite the distance to the node through which the route indicated by the one shown in FIG. 2 is changed to that of the route via the shortest route and the route indicated by the information, and add the upstream / downstream relationship corresponding to the downstream of the relay node to the calculation target tree. Is handled as a node through which the computation object tree passes .

The shortest path calculation method of the present invention is a method for calculating the shortest path between nodes in a network,
When there is a change in the topology of the network, information including nodes that can be reached by following the upstream / downstream relationship between the nodes from the source node that is the source node and the upstream / downstream connection relationship between the nodes is calculated. Tree and
When the topology change is the addition of a new first link, the first and second nodes that are nodes in the calculation target tree and are connected by the first link are referred to from the origin node to the A first path comprising a path reaching the second node along the calculation target tree and a path reaching the first node via the path of the first link; and the calculation from the origin node Comparing the distance with the second path that reaches the first node along the target tree, and if the distance of the first path is shorter than the second path, the first path is determined as the starting point. A shortest path candidate from a node to the first node,
If the topology change is link deletion, information on the downstream side of the deleted link is deleted from the calculation target tree, and a third node in the calculation target tree and a fourth that does not belong to the calculation target tree When the second link connecting the node to the second node is found, a route that reaches the third node along the tree to be calculated from the starting node and a route that includes the second link are A candidate for the shortest route to the node,
For a route from the origin node to the first or fourth node, if there are a plurality of candidates for the shortest route, a candidate having the shortest distance is identified from a plurality of candidates, and the identified candidate is determined as the shortest route ,
If there is information corresponding to an upstream / downstream relationship between nodes corresponding to a downstream path that is opposite to the origin node for the relay node that is the first or the fourth node, the information Rewrite the distance to the node through which the route indicated by the one shown in FIG. 2 is changed to that of the route via the shortest route and the route indicated by the information, and add the upstream / downstream relationship corresponding to the downstream of the relay node to the calculation target tree. Is treated as a node through which the computation object tree passes .

  Furthermore, a program according to the present invention causes a computer to execute the shortest path calculation method according to the present invention.

  According to the present invention, it is possible to calculate the shortest path even if a topology changes due to a new link being added to the network or a link being deleted from the network. Further, even when there is a newly added link that cannot be calculated by the method disclosed in Non-Patent Document 2, the shortest path can be calculated.

  The configuration of the shortest path calculation apparatus according to this embodiment will be described. FIG. 1 is a block diagram illustrating a configuration example of the shortest path calculation apparatus according to the present embodiment.

  As shown in FIG. 1, a shortest path calculation device 10 of this embodiment includes a storage unit 12 that stores information necessary for route calculation, a processor 14 including an information management unit 141 and a route calculation unit 142, and an external device. And an interface unit 16 for transmitting and receiving data.

  The storage unit 12 stores path information, which is information on the shortest path between nodes in the network, and topology information indicating the connection status of physical links and virtual links between nodes in the network. The storage unit 12 also stores information during calculation until the shortest path is determined.

  The interface unit 16 notifies the calculation result by the processor 14 to another device via a network (not shown), and passes information on the topology received from the other device via the network to the processor 14.

  The processor 14 is provided with a memory (not shown) for storing a program. When the processor 14 executes the program, the information management unit 141 and the route calculation unit 142 are virtually configured in the shortest route calculation device 10.

  The information management unit 141 monitors the network status, and when a link is newly set in the network or a link in the network is deleted, notifies the route calculation unit 142 of information on such changes. Further, when the calculation result is acquired from the route calculation unit 142 or the information about the topology is acquired from the interface unit 16, the information in the storage unit 12 is updated based on the acquired calculation result and information. As an example of a method of monitoring the network status, there is a method of acquiring a routing table exchanged between routers, but other methods may be used.

  When the route calculation unit 142 receives information on changes in the network status from the information management unit 141, the route calculation unit 142 performs route calculation and notifies the information management unit 141 of the calculation result. Details of the route calculation method will be described in the procedure of the shortest route calculation method described later.

  As described above, the shortest path calculation device 10 has a path calculation function, receives a routing table of each router in the network, or takes a route calculation of each router at once, and each router in the network based on the calculation result. Is a computer that instructs the setting of the output destination. The shortest path calculation apparatus 10 may be a specific router in the network, or may be an apparatus that specially executes path calculation processing.

  Consider an example of calculating the shortest path connecting two nodes in a network using this shortest path calculation device.

  The shortest path calculation method of the present embodiment is a method of calculating the shortest path with a plurality of other nodes at a time from one node among the shortest paths connecting two nodes. Here, terms used in the present embodiment will be described.

  2A and 2B are diagrams illustrating an example of information indicating a route starting from one node. 2A is a diagram showing the shortest path in a tree structure, and FIG. 2B is a diagram showing the shortest path individually.

  The “route tree” indicates a part of routes that have the same starting point among routes connecting two points. In the present embodiment, a route set that is a set of routes is represented by a tree structure. A path connecting two nodes consists of a node and a link that pass through the tree from the start point to the end point. If a tree as shown in FIG. 2A exists, the node s is the starting point, and the path from the node s to b is the node that passes through the tree from the node s to b, s, a, It is expressed as s → a → b using b.

  The node s is on the upstream side when viewed from each of the nodes a and d, and the node a is on the upstream side when viewed from each of the nodes b and c. Conversely, the nodes a to d are on the downstream side when viewed from the node s.

  Some routes included in the set have a common part. As shown in FIG. 2B, it is redundant in terms of information to have these pieces of information individually. By representing in a tree structure, redundant information can be collected and represented by information for managing common parts. In the tree shown in FIG. 2A, a path connecting nodes s to b and nodes s to c passes between nodes s and a. In this case, instead of mentioning passing between nodes s and a by an individual route (see FIG. 2B), if the tree structure is as shown in FIG. 2A, there is a route passing between nodes s and a. It is possible to indicate that each passes between the nodes s and a. Thereby, it can prevent having redundant information. In this example, a link connecting nodes s → a in the tree is used as information indicating the existence of the shortest path.

  The information of the tree structure shown in FIG. 2A becomes the route information stored in the storage unit 12. Although not shown in the figure, routes connecting nodes are listed as candidate routes, and only the route that finally becomes the shortest route is recorded in the storage unit 12 as route information, leaving the tree structure shown in FIG. 2A.

  “Subordinate tree of node Z” refers to a link and a node located downstream from node Z in the information of the path tree passing through node Z. Node Z itself is also included in the subordinate tree. Using the example of the path tree in FIG. 2A, the subordinate tree of the node a is a part connecting the nodes a, b, and c in the path tree.

  The “distance” between nodes does not mean the actual length between nodes of the network, but the total sum of the costs assigned to the links on the route. If the link cost is the same between any nodes, the “distance” is proportional to the number of hops.

  The procedure of the shortest path calculation method executed by the shortest path calculation apparatus 10 of this embodiment will be described. 3 and 4 are flowcharts showing the procedure of the shortest path calculation method of the present embodiment.

  Here, it is considered that the shortest path tree S′x is a calculation target in the shortest path set starting from the node X. It is assumed that information of the shortest path set starting from the node X before the topology change is registered in advance in the storage unit 12 as the shortest path tree Sx.

  As shown in FIG. 3, in the first procedure, the shortest path calculation apparatus 10 monitors the network (step 101), and checks whether there is a link deletion or a link addition (step 102). If there is no link deletion or addition, the process returns to step 101.

  Consider a case where a newly set link L1 exists in the network topology at nodes Ea and Eb through which the calculation target tree S'x passes. In step 102, when the shortest path calculation device 10 recognizes that the link L1 is newly set in the network, it calculates the distance to the node Ea for the following two paths. The first route is a combination of the route from the node X to the node Eb in the calculation target tree S′x (this route is referred to as route R2) and the route from the node Eb to the node Ea via the link L1. The route is (R2 + L1). The second path is a path from the node X to the node Ea in the calculation target tree S′x (this is referred to as a path R1). The shortest path calculation device 10 compares the path (R2 + L1) and the path R1, and if the former distance is shorter, the former path is treated as a candidate path (step 103).

  On the other hand, consider the case where the link has been deleted from the network. In step 102 or 104, when the shortest path calculation device 10 recognizes that the link has been deleted from the network, the shortest path tree Sx corresponds to the upstream / downstream relationship across the node or link deleted from the topology. Check if the information exists. If there is a node or link to be deleted, information on the upstream / downstream relationship between the corresponding nodes is deleted. If there are no nodes or links to be deleted, leave them as they are. Including the nodes that can be reached from the node X by following the path tree and the upstream / downstream relationship between the nodes, these are treated as the shortest path tree S'x to be calculated (step 105).

  Subsequently, the shortest path calculation device 10 searches for a link L2 connecting the node Ec through which the calculation target tree S′x passes and the node Ed through which the calculation target tree S′x does not pass. When the link L2 is found as a result of the search, the route reaching the node Ed from the node X reaches the node Ed via the calculation target tree S′x and the node Ec via the link L2. Attention is paid to a route that is combined with the route to be performed (this route is referred to as route C). Then, this route C is treated as one of the candidate routes (step 106). If a link is added, the process proceeds to step 103 (step 107).

  As described above, in both cases where a link is newly added and a link is deleted, the shortest path calculation device 10 finds candidate paths and repeats the following operations until there are no candidate paths. .

  If there are a plurality of candidate routes (step 108), the shortest route calculation device 10 selects a candidate route having the shortest distance. If the selected route is the shortest among the routes to the end point of the known route up to the selected time, the selected candidate route is set as the shortest route to the end point of the candidate route, that is, the calculation result (step 109). ).

  When the node Ea when the link is newly set and the node Ed when the link is deleted are referred to as a node E1 below, the shortest path calculation device 10 is a node E1 in the calculation target tree S′x. Change the route information up to that of the selected route. Specifically, the upstream / downstream relationship between the nodes is changed according to the selected route (step 110 shown in FIG. 4). Note that the node E1 corresponds to the relay node of the present invention.

  If there is a subordinate tree at the node E1, the shortest path calculation device 10 rewrites the distance to the node passing through the subordinate tree to a combination of the path selected at step 109 and the path via the subordinate tree (step 111). ). As a result, the distance of the path from the starting point to the node E1 and the distance from the node E1 to the node constituting the subordinate tree have changed, so that in some nodes, the subordinate tree is more than the current path found at this point. In some cases, the route passing through the nodes constituting the node becomes shorter. In addition, when a link newly added to the network topology connects nodes constituting the subordinate tree, the route via the newly added link is shorter than the route via the shortest route to the subordinate tree and the node E1. It also occurs. In order to search for such a route, the shortest route calculation apparatus 10 performs the following operations of Step 112 and Step 113.

  In step 112, the shortest path calculation device 10 searches for a link connecting nodes constituting the subordinate tree of the node E1 among the links newly added to the topology. If there is a link L3 connecting the nodes c1 and c2 constituting the subordinate tree of the node E1, the end point c1 is reached via the shortest path from the node X to the node E1 at the starting point of the tree, the subordinate tree, and the link L3. The distance of the route (referred to as route R5) is compared with the distance of the shortest route from node X to node E1 and the route (referred to as route R4) reaching the end point c1 via the subordinate tree. If there is an end point at which the route R5 is shorter than the route R4, a process for handling the route R5 as a candidate route is performed.

  In step 113, the shortest path calculation device 10 searches for a link starting from a node constituting the subordinate tree of the node E1 and ending at a node not constituting the subordinate tree. If there is a link L4 starting from the node d1 of the subordinate tree of the node E1 and ending with the node d2 that does not constitute the subordinate tree, the shortest path from the node X to the node E1 and the path from the node E1 to the end point d1 of the tree And a path including the shortest path from the node X to the node E1 and the path from the node E1 to the end point d2 (referred to as a path R6). ) And the distance. Then, if there is an end point such that the route R7 is shorter than the route R6, processing for treating the route R7 as a candidate route is performed. When the operations in steps 112 and 113 are completed, the process returns to step 108.

  The procedure of the shortest path calculation method of this embodiment will be described using a specific example. Here, while referring to FIGS. 3 and 4, the operation of each step will be described with reference to FIGS. 5 to 10.

  FIG. 5 is a diagram illustrating the shortest path tree Sx that starts from the node X and is calculated before the topology change. Here, in addition to the node X as a starting point, nodes A to D as representative nodes in the following description are shown in the figure. It should be noted that broken line arrows indicating the shortest paths between nodes may be routed through other nodes and are not shown in the figure.

  FIG. 6 is a diagram for explaining the processing of step 103 shown in FIG.

  When a new link is set in the network, in step 103, the shortest path calculation device 10 searches for a new additional link L1 that connects nodes through which the calculation target tree S'x passes. When the link L1 is found, the path R1 (X → J → G or X → H) that has been found so far among the paths that reach the node G and the node H, which are the end points, and S′x And the route R2 (X → H → G or X → J → G → H) passing through L1. If the route R2 is shorter than the route R1, the route R2 is treated as a candidate route.

  Paying attention to the route to the node G, the shortest route calculation device 10 compares the route R1 (X → J → G) and the route R2 (X → H → G), and the distance of the route R2 is shorter than the route R1. For example, route R2 is treated as a candidate route. FIG. 6 shows a case where attention is paid to the route to the node G. The route to the node H is performed in the same manner as the node G.

  FIG. 7 is a diagram showing an example when a link is deleted on the network. Although FIG. 7 shows a case where two links are deleted, one link may be deleted, or three or more links may be deleted. FIG. 7 shows the processing of step 105 shown in FIG. FIG. 8 is a diagram for explaining the processing of step 106 shown in FIG. 3, and FIG. 9 is a diagram for explaining the processing of step 109 shown in FIG.

  As shown in FIG. 7, it is assumed that the link A → B connecting the nodes A and B and the link C → D connecting the nodes C and D are deleted from the topology. The shortest path calculation device 10 has an upstream / downstream relationship across the links A → B and C → D deleted in the shortest path tree Sx, that is, a connection relation between the node A and the node B, and the node C and the node D. Check if there is a connection relationship between them. In this example, both exist. At this time, the shortest path calculation device 10 deletes information corresponding to the links A → B and C → D from the shortest path tree Sx. By this operation, the shortest path tree Sx is divided into a path tree connecting node A and node C with node X as a vertex, and a path tree with nodes B and D as vertices, respectively. A partial tree connecting from the node X to the nodes A and C that can be reached from the node X via the tree is defined as a calculation target tree S′x.

  In step 106, the shortest path calculation device 10 searches for a path to a node constituting the calculation target tree S′x divided in step 105. In this algorithm, it is possible to reach the end point of a route from the node in the calculation target tree S′x to the outside of the calculation target tree S′x with one hop (link L2), and the calculation target tree S′x is reached to the start point of the link L2. A route to be taken is set as a search target. In FIG. 8, the route X → N → P from the node X to the node P corresponds to such a route. The found route is treated as a candidate for the shortest route from node X to node P.

  The shortest path calculation device 10 treats each of the cases where a link is newly set and when the link is deleted, and is treated as a candidate path in step 109 shown in FIG. Among the routes, the route having the shortest distance to the end point is selected.

  In the case where the link is deleted, the shortest path calculation apparatus 10 as shown in FIG. 9 will be used if the selected path is the shortest among the paths to the end point of the path known up to the selected time point. The selected route is treated as the shortest route R3 connecting the starting node X and the ending node P of the calculation target tree. Subsequently, in step 110 shown in FIG. 4, the information related to the upstream node in the end point node P of the shortest path R3 is rewritten to the information related to the node N upstream in the shortest path R3. That is, the parent node of the node P is rewritten from the node Q to the node N from FIGS.

  The following description focuses on the node P when the link is deleted on the network. However, when a new link is set, the node P may be replaced with the node G. Description is omitted.

  If there is a node on the subordinate tree of the end node P of the route, the shortest path calculation device 10 calculates the distance to the node on the tree by the route R3 and the route reaching the node on the tree through the subordinate tree. Change to distance. In the example shown in FIG. 9, there are a node S and a node T under the node P. If subordinate tree information exists in the storage unit 12, the distance from the subordinate tree to each node can be easily calculated from the tree information.

  If the distance from the node P to the node T existing on the subordinate tree is 2 and the distance from the starting node X to the node P of the calculation target tree S′x is 5, the shortest path R3 and the subordinate tree are passed. The distance of the route to be calculated is 7. The value of the distance 7 is a distance value from the node X to the node T. Then, the subordinate tree of the node P is included in the calculation target tree S′x.

  FIG. 10 is a diagram for explaining the processing of step 112 and step 113 shown in FIG.

  In step 112, the shortest path calculation device 10 searches for a link connecting nodes existing on the subordinate tree of the node P among the links newly added to the topology. Of the two ends of the link L3 (node S, node T), the route through the calculation target tree S′x and the link L3 rather than the path reaching the nodes via the calculation target tree S′x. Find a route that has a shorter distance. In the example in FIG. 10, in order to reach the node T existing on the subordinate tree of the node P, the route R4 (X → P → T) via the calculation target tree S′x and the route R5 (X Compare the distance of P → S → T). Then, the case where the distance of the route R5 is shorter than the route R4 is searched. If such a case exists, the route R5 is treated as one of candidate routes.

  In step 113, the shortest path calculation device 10 searches for a path having a shorter distance than the path existing in the calculation target tree S′x via the subordinate tree of the node P. At this time, a route to the end node where the distance from the node on the subordinate tree is known is searched.

  Among the applicable routes, this algorithm can reach the end point of the route with one hop (link L4), and the route that passes through the calculation target tree S′x to the start point of the link is the search target. In FIG. 10, if there is a link L4 starting from the node S of the subordinate tree of the node P and ending at the node M that does not constitute the subordinate tree, the shortest path from the node X to the node P at the starting point of the tree and the node P to the node The distance to the node M via the path to S and the link L4 (path R7: path X → P → S → M), and the node M from the node X via the calculation target tree S′x And the distance of the shortest path candidate (referred to as the path R6) to be compared. Then, if there is an end point such that the route R7 is shorter than the route R6, processing for treating the route R7 as a candidate route is performed. The found route R7 is treated as one of the candidate routes.

  In FIG. 10, the route R6 is expressed as X → node (not shown) → M, but the route R6 may be X → P → M. Various routes are conceivable as the route R6, and FIG. 10 shows only one example.

  The shortest path calculation apparatus 10 repeats the operations of step 112 and step 113 until there is no path that is treated as a candidate path (step 108).

  According to the present invention, it is possible to calculate the shortest path even if a topology changes due to a new link being added to the network or a link being deleted from the network. Further, even when there is a newly added link that cannot be calculated by the method disclosed in Non-Patent Document 2, the shortest path can be calculated. This is because a route that passes through a newly added link is also a search target. This makes it possible to calculate the shortest path even when a link is added.

It is a block diagram which shows the example of 1 structure of the shortest path calculation apparatus of this embodiment. It is the figure which represented the shortest path | route with the tree structure about the path | route which starts from one node. It is the figure which represented the shortest path | route separately about the path | route which starts from one node. It is a flowchart which shows the procedure of the shortest path | route calculation method of this embodiment. It is a flowchart which shows the procedure of the shortest path | route calculation method of this embodiment. It is a figure which shows the shortest path | route tree Sx calculated from the node X calculated before topology change. It is a figure for demonstrating the process of step 103 shown in FIG. It is a figure which shows an example when the link is deleted by the network. It is a figure for demonstrating the process of step 105 and step 106 shown in FIG. It is a figure for demonstrating the process of step 109 shown in FIG. It is a figure for demonstrating the process of step 112 shown in FIG. 4, and step 113. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Shortest path calculation apparatus 12 Memory | storage part 14 Processor 16 Interface part 141 Information management part 142 Path | route calculation part

Claims (7)

A shortest path calculation device that calculates the shortest path between nodes in a network,
A storage unit for storing a calculation target tree including information on a node that can be reached from a starting point node that is a starting point node by following an upstream / downstream relationship between the nodes, and an upstream / downstream connection relationship between the nodes;
When there is a change in the topology of the network, the calculation target tree is set. When the change in the topology is addition of a new first link, the first link is a node in the calculation target tree. For the first and second nodes connected in the above, the first node via the path from the origin node to the second node along the calculation object tree and the path of the first link The distance between a first route that is a route that reaches the first node and a second route that is reached from the origin node along the calculation target tree to the first node is compared, and the distance from the second route is more than the second route. If the distance of the first route is short, the first route is determined as a candidate for the shortest route from the origin node to the first node, and the deletion is performed when the change in topology is deletion of a link. When the information on the downstream side of the calculated link is deleted from the calculation target tree, and the second link connecting the third node in the calculation target tree and the fourth node not belonging to the calculation target tree is found, A route consisting of the second link and a route reaching the third node along the calculation target tree from the origin node is taken as a candidate of the shortest route from the origin node to the fourth node, and A processor for identifying a candidate having the shortest distance from a plurality of candidates and determining the identified candidate as the shortest path if there are a plurality of candidates for the shortest path with respect to the route to the first or fourth node;
I have a,
The processor is
If there is information corresponding to an upstream / downstream relationship between nodes corresponding to a downstream path that is opposite to the origin node for the relay node that is the first or the fourth node, the information Rewrite the distance to the node through which the route indicated by the one shown in FIG. 2 is changed to that of the route via the shortest route and the route indicated by the information, and add the upstream / downstream relationship corresponding to the downstream of the relay node to the calculation target tree A shortest path calculation device that treats a node positioned at a node as a node through which the calculation object tree passes . The processor is
Among the links newly added to the topology, if there is a third link connecting nodes located downstream from the relay node that is the first or fourth node, the end point of the third link The starting point comprising the shortest path from the starting node to the relay node, the path to the sixth node downstream of the relay node, and the third link for a fifth node and a sixth node A fourth path comprising a third path from a node to the fifth node, the shortest path from the origin node to the relay node, and a path to the fifth node downstream of the relay node If the distance of the third route is shorter than the fourth route by comparing the distance with the route, the shortest distance from the origin node to the fifth node is the third route. A candidate route of the road, pick out what distance of the candidate route is the shortest, shortest path calculation apparatus according to claim 1. The processor is
A seventh node located downstream of the relay node that is the first or fourth node is a starting point, and an eighth node that is not located downstream of the relay node and for which the shortest path has not been calculated is an end point. The fourth link to be configured includes the path from the origin node to the relay node via the shortest path to the seventh node downstream of the relay node and the fourth link. Compare the distance between the fifth route from the node to the eighth node and the sixth route from the origin node to the eighth node via the computation object tree. If the distance of the fifth route is shorter than the sixth route, the fifth route is set as the shortest candidate route from the origin node to the eighth node, and the distance among the candidate routes is The shortest Out beauty, the shortest path calculating apparatus according to claim 2 wherein. A method for calculating the shortest path between nodes in a network,
When there is a change in the topology of the network, information including nodes that can be reached by following the upstream / downstream relationship between the nodes from the source node that is the source node and the upstream / downstream connection relationship between the nodes is calculated. Tree and
When the topology change is the addition of a new first link, the first and second nodes that are nodes in the calculation target tree and are connected by the first link are referred to from the origin node to the A first path comprising a path reaching the second node along the calculation target tree and a path reaching the first node via the path of the first link; and the calculation from the origin node Comparing the distance with the second path that reaches the first node along the target tree, and if the distance of the first path is shorter than the second path, the first path is determined as the starting point. A shortest path candidate from a node to the first node,
If the topology change is link deletion, information on the downstream side of the deleted link is deleted from the calculation target tree, and a third node in the calculation target tree and a fourth that does not belong to the calculation target tree When the second link connecting the node to the second node is found, a route that reaches the third node along the tree to be calculated from the starting node and a route that includes the second link are A candidate for the shortest route to the node,
For a route from the origin node to the first or fourth node, if there are a plurality of candidates for the shortest route, a candidate having the shortest distance is identified from a plurality of candidates, and the identified candidate is determined as the shortest route ,
If there is information corresponding to an upstream / downstream relationship between nodes corresponding to a downstream path that is opposite to the origin node for the relay node that is the first or the fourth node, the information Rewrite the distance to the node through which the route indicated by the one shown in FIG. 2 is changed to that of the route via the shortest route and the route indicated by the information, and add the upstream / downstream relationship corresponding to the downstream of the relay node to the calculation target tree A shortest path calculation method in which a node located at is treated as a node through which the calculation object tree passes . Among the links newly added to the topology, if there is a third link connecting nodes located downstream from the relay node that is the first or fourth node, the end point of the third link The starting point comprising the shortest path from the starting node to the relay node, the path to the sixth node downstream of the relay node, and the third link for a fifth node and a sixth node A fourth path comprising a third path from a node to the fifth node, the shortest path from the origin node to the relay node, and a path to the fifth node downstream of the relay node Compare the distance to the route,
If the distance of the third route is shorter than the fourth route, the third route is set as a candidate route of the shortest route from the origin node to the fifth node, and the distance is shortest among the candidate routes. 5. The shortest path calculation method according to claim 4 , wherein the shortest path is selected. A seventh node located downstream of the relay node that is the first or fourth node is a starting point, and an eighth node that is not located downstream of the relay node and for which the shortest path has not been calculated is an end point. The fourth link to be configured includes the path from the origin node to the relay node via the shortest path to the seventh node downstream of the relay node and the fourth link. Compare the distance between the fifth route from the node to the eighth node and the sixth route from the origin node to the eighth node via the computation object tree. ,
If the distance of the fifth route is shorter than the sixth route, the fifth route is set as a candidate route of the shortest route from the origin node to the eighth node, and the shortest distance among the candidate routes is 6. The shortest path calculation method according to claim 4, wherein a shortest path is selected. Program for executing one of a shortest path calculating method of claims 4 6 to the computer.

Images