WO2024116350A1 - Path calculation device, program and path calculation method - Google Patents

Abstract

This path calculation device (100) comprises: a minimum-cost path calculation unit (113) that calculates a minimum-cost path from a starting point node to an end point node in a communication network (refer to a network 500) configured by including nodes as well as links to which costs are assigned, said communication network being further configured such that a packet, to which a label sequence, i.e., the sequence of labels indicating nodes through which the packet passes while being transferred is assigned, is so transferred as to minimize the cost at which the pack passes through the nodes indicated by the labels included in the label sequence; and a label calculation unit (114) that determines one minimum cost path and calculates the label sequence of the minimum cost path for which the number of labels is smaller than the number of nodes included in the determined minimum cost path.

Description

Route calculation device, program, and route calculation method

The present invention relates to a route calculation device, program, and route calculation method for calculating labels to be assigned to packets in a communication network where packets are forwarded using labels.

The spread of fifth-generation mobile communication systems (5G) and esports has created demands for strict network requirements, such as ultra-low latency and low jitter. To ensure these requirements, network monitoring and control technologies are needed. Existing monitoring technologies can measure latency, jitter, and packet loss on any route or link by connecting a measurement system to one point on the network and using a path control mechanism based on traffic engineering (Patent Documents 1-3).

International Publication No. 2021/166266 International Publication No. 2021/166267 International Publication No. 2022/176123

In the techniques described in Patent Documents 1 to 3, in a network where label-based packet forwarding (e.g., Multi-Protocol Label Switching) is performed, all labels indicating relay nodes are assigned to the measurement packet. Therefore, in a large-scale network, the number of labels assigned to the measurement packet may exceed the specifications that the relay node can process. Therefore, it is necessary to reduce the number of labels and calculate labels that allow the measurement packet to pass through the route to be measured. In addition, a large-scale network is generally constructed from multiple domains, and it is necessary to calculate the measurement route and labels with consideration of the domain boundaries.
The present invention has been made in view of the above background, and an object of the present invention is to enable packet forwarding through a desired route without attaching labels of all relay nodes.

In order to solve the above-mentioned problems, the route calculation device according to the present invention is a communication network including nodes and links to which costs are assigned, in which packets to which a label string, which is a string of labels indicating the nodes that the packets will pass through during forwarding, are forwarded so as to minimize the cost by passing through the nodes indicated by the labels included in the label string. The device includes a minimum-cost route calculation unit that calculates a minimum-cost route from a start node to a destination node in the communication network, and a label calculation unit that identifies one of the minimum-cost routes and calculates a label string for the minimum-cost route with a number of labels less than the number of nodes included in the identified minimum-cost route.

This invention makes it possible to forward packets along a desired route without assigning labels to all relay nodes.

1 is a functional block diagram of a path calculation device according to an embodiment of the present invention. FIG. 1 is a configuration diagram of a network used to explain this embodiment. FIG. 2 is a diagram illustrating a network configuration in which boundary nodes are arranged by domain and are used to explain the present embodiment. 11 is a flowchart of a label calculation process according to the embodiment. 13 is a flowchart of a label calculation process for a domain-specific route according to a modified example of the present embodiment. FIG. 2 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the path calculation device according to the embodiment described above.

Below is a description of a route calculation device that calculates routes within a network in a form (embodiment) for implementing the present invention. This network forwards packets so that they pass through relay nodes indicated by the labels attached to the packets. For the portions of the route that do not have labels attached, the packets are forwarded so that they pass through the route with the smallest cost.

The route calculation device calculates the minimum cost route between the start node and end node of a route (measurement route) within a network that is the subject of measurement of delay, jitter, etc. Next, the route calculation device calculates a label string including labels indicating relay nodes on the route. The route calculation device calculates the label string so that the measurement route is the only minimum cost route that passes through the calculated label string, and the number of labels is small. In addition, when the network is composed of multiple domains, the route calculation device calculates the label string so that the measurement route passes through the boundary nodes of the domains.

With this type of route calculation device, it is possible to measure the delay and jitter of the measurement route using measurement packets with a reduced number of labels. This makes it possible to measure even large-scale networks that could not be measured using the labels of all relay nodes in the past. Furthermore, this route calculation device has the same effect in multi-domain networks.

<Configuration of the route calculation device>
1 is a functional block diagram of a path calculation device 100 according to this embodiment. The path calculation device 100 is a computer, and includes a control unit 110, a storage unit 120, and a communication unit 180. The communication unit 180 includes a communication device, and is connected to a network 500 including nodes.

<Route calculation device: storage unit>
The storage unit 120 includes storage devices such as a read only memory (ROM), a random access memory (RAM), a solid state drive (SSD), etc. The storage unit 120 stores a network information database 130, a measurement result database 140, and a program 128.

The network information database 130 stores information related to the nodes and links included in the network 500, such as node and link identification information, connection information, and link costs. When the network 500 is made up of multiple domains (see FIG. 2 described later), the network information database 130 stores information on the domains to which the nodes belong. Note that nodes that belong to multiple domains are called boundary nodes. When a packet is sent between two domains, the boundary node transfers the packet from one domain to the other domain.
The measurement result database 140 stores the measurement results of nodes, links, and routes.
The program 128 includes a description of the procedure for the label calculation process (see FIG. 4) described later.

<Route calculation device: control unit>
The control unit 110 is configured to include a CPU (Central Processing Unit), and is provided with a network information collection unit 111 , a measurement unit 112 , a minimum cost path calculation unit 113 , and a label calculation unit 114 .

<Control unit: Network information collection unit>
The network information collection unit 111 acquires connection information (topology information) between nodes and links in the network 500 and link costs, and stores the information in the network information database 130. The network information collection unit 111 acquires topology information and link costs by acquiring information on OSPF (Open Shortest Path First) and BGP-LS (Border Gateway Protocol - Link State) from the nodes in the network 500, for example.

<Control unit: Measurement unit>
The measurement unit 112 uses measurement packets to measure the performance of nodes, links, and paths between two nodes, and stores the results in the measurement result database 140. For example, the measurement unit 112 measures delay and jitter on a path between two nodes to be measured. There is not necessarily only one minimum-cost path between two nodes. There may be multiple minimum-cost paths, and each minimum-cost path is measured. For this reason, it is required that the measurement packet be given labels of the nodes through which the measurement packet passes, so that the measurement packet passes (is forwarded) along each minimum-cost path.

<Control unit: Minimum cost route calculation unit>
The minimum cost path calculation unit 113 calculates a minimum cost path between a start node and an end node, which is a path with a minimum sum of costs of links on the path. The path is represented by a sequence of nodes. There may be more than one minimum cost path between a start node and an end node. Furthermore, when the start node and the end node are in different domains, the minimum cost path calculation unit 113 calculates a minimum cost path that passes through a boundary node.

<Control unit: Label calculation unit>
The label calculation unit 114 calculates a label string to be assigned to the packet so that the packet is forwarded through a measurement path within the network 500. This label string will be described below.
Packets are transferred through network 500 in the order of one or more labels (label string) assigned to the packet, passing through nodes indicated by the labels. Between a node indicated by one label in the label string and a node indicated by the next label, the packet is transferred through a route with the minimum cost. In other words, for each label included in the assigned label string, the packet is transferred through a minimum-cost route between the node indicated by the label and the node indicated by the next label in the label string. If there are multiple minimum-cost routes between a node indicated by a label and a node indicated by the next label, it is undefined which minimum-cost route the packet will take.

The label calculation unit 114 calculates a label string to be assigned to the measurement packet so that the route taken is not undefined, the measurement route is followed, and the number of labels is smaller than when all nodes passed through are specified. Note that the label string is calculated to include the start node, end node, and boundary node. The label calculation method used by the label calculation unit 114 will be described later with reference to FIG. 4.

<Network>
2 is a configuration diagram of a network 500 used to explain this embodiment. The network 500 is a multi-domain network made up of domains A, B, and C. Domain A is made up of nodes 1 to 7. Domain B is made up of nodes 6 to 11, and domain C is made up of nodes 10 to 14. Nodes 6 and 7 are boundary nodes between domains A and B, and nodes 10 and 11 are boundary nodes between domains B and C.

The cost of a link connecting nodes is indicated by the width of the line connecting the nodes; the wider the line, the lower the cost. In Figure 2, a wide link has a cost of 1, and a narrow link has a cost of 10. For example, the cost of the route between nodes 6 and 9 that passes through nodes 6, 8, and 9 is 2, and the cost of the route that passes through nodes 6, 7, and 9 is 20, so the minimum cost route is the route that passes through nodes 6, 8, and 9.

FIG. 3 is a configuration diagram of a network 500 in which boundary nodes used in the description of this embodiment are shown by domain. In comparison with FIG. 2, in FIG. 3, the boundary nodes 6, 7, 10, and 11 are shown as different nodes by domain. Nodes 6 and 6X are the same node, but are shown as node 6 in domain B and node 6X in domain A. Nodes 7 and 7X are the same node, but are shown as node 7 in domain B and node 7X in domain A. Nodes 10 and 10X are the same node, but are shown as node 10 in domain B and node 10X in domain C. Nodes 11 and 11X are the same node, but are shown as node 11 in domain B and node 11X in domain C. The cost of the links between nodes 6 and 6X, between nodes 7 and 7X, between nodes 10 and 10X, and between nodes 11 and 11X is considered to be 0.

<Label sequence of minimum cost route>
There are four minimum cost routes from node 1 to node 14 passing through boundary nodes: a route passing through nodes 1, 2, 3, 6X, 6, 8, 9, 11, 11X, 12, 14, a route passing through nodes 1, 4, 5, 6X, 6, 8, 9, 11, 11X, 12, 14, a route passing through nodes 1, 2, 3, 6X, 6, 8, 9, 11, 11X, 13, 14, and a route passing through nodes 1, 4, 5, 6X, 6, 8, 9, 11, 11X, 13, 14. Minimum cost route calculation unit 113 calculates these four routes as minimum cost routes with node 1 as the start node and node 14 as the end node.

The label calculation unit 114 calculates a label string that has fewer labels than when all nodes passed through for one minimum-cost route are specified. In the following explanation, the minimum-cost route is a route that passes through nodes 1, 2, 3, 6X, 6, 8, 9, 11, 11X, 12, and 14 (hereinafter referred to as the specified route). Note that in the following explanation, nodes and labels are considered to be the same. For example, the label of node 1 is 1.

For example, if a label string of 1, 2, 6X, 11, 12, 14 is specified, the minimum cost route passing through nodes 1, 2, 6X, 11, 12, 14 is the only specified route. In contrast, if a label string of 1, 2, 6X, 11, 14 is specified, the minimum cost route passing through nodes 1, 2, 6X, 11, 14 includes a route that passes through nodes 1, 2, 3, 6X, 6, 8, 9, 11, 11X, 13, 14 in addition to the specified route, and the route is not specified (specified), and the route becomes undefined. The label calculation unit 114 calculates a label string that has fewer labels than when all nodes are specified, and results in a single minimum cost route (the route is not undefined).

<Label calculation process>
Fig. 4 is a flowchart of the label calculation process according to this embodiment. With reference to Fig. 4, the process will be described in which the minimum cost path calculation unit 113 calculates one or more minimum cost paths for the start node and end node specified by the measurement unit 112, and the label calculation unit 114 calculates a label (label string) for each minimum cost path. In addition to the description of the label calculation process itself, a specific example will be described in which the start node is node 1 and the end node is node 14.

In step S11, the minimum cost path calculation unit 113 calculates the minimum cost path from the start node to the end node. In a specific example, four paths are calculated: a path passing through nodes 1, 2, 3, 6X, 6, 8, 9, 11, 11X, 12, 14; a path passing through nodes 1, 4, 5, 6X, 6, 8, 9, 11, 11X, 12, 14; a path passing through nodes 1, 2, 3, 6X, 6, 8, 9, 11, 11X, 13, 14; and a path passing through nodes 1, 4, 5, 6X, 6, 8, 9, 11, 11X, 13, 14.

In step S12, the label calculation unit 114 starts the process of repeating steps S13 to S19 for each route (minimum cost route) calculated in step S11. In the following, the route that is the target of this repeated process is referred to as the target route. In the following, of the four minimum cost routes, the route that passes through nodes 1, 2, 3, 6X, 6, 8, 9, 11, 11X, 12, and 14 will be described as a specific example.

In step S13, the label calculation unit 114 divides the target route by domain. In a specific example, the target route is divided into a route passing through nodes 1, 2, 3, and 6X, a route passing through nodes 6, 8, 9, and 11, and a route passing through nodes 11X, 12, and 14 (hereinafter, referred to as "domain-specific routes").
In step S14, the label calculation unit 114 starts the process of repeating steps S15 to S18 for each of the domain-specific routes divided in step S13. Hereinafter, the domain-specific route that is the target of this repeated process is referred to as a target domain-specific route.

In step S15, the label calculation unit 114 calculates a label string in which intermediate nodes between the start node and the end node of the target domain-specific route are deleted. As a specific example, in the case of a route passing through nodes 1, 2, 3, and 6X, a label string of 1, 6X is calculated.
In step S16, the label calculation unit 114 calculates the minimum cost route that passes through the nodes indicated by the label string calculated in step S15.

In step S17, if the number of minimum cost routes calculated in step S16 is one (step S17→YES), the label calculation unit 114 returns to step S15 to process the next domain-specific route, and if the number of minimum cost routes is not one (step S17→NO), the label calculation unit 114 proceeds to step S18.
In step S18, the label calculation unit 114 adds one label to the label string. For example, the label of the first node not included in the label string is added, and the process returns to step S16.

Here is a concrete example. In step S16, there are two minimum cost routes indicated by the label string 1, 6X: one that passes through nodes 1, 2, 3, 6X, and one that passes through nodes 1, 4, 5, 6X. In step S17, there are two routes: one that passes through nodes 1, 2, 3, 6X, and one that passes through nodes 1, 4, 5, 6X, so the program proceeds to step S18. As the node that is not included in the label string of 1, 6X on the route that passes through nodes 1, 2, 3, 6X is node 2, 2 is added to the label string, resulting in the label string 1, 2, 6X, and the program returns to step S16.

The minimum cost route indicated by the label sequence of 1, 2, 6X is the only route that passes through nodes 1, 2, 3, and 6X, so after processing the per-domain route for domain A, the label calculation unit 114 executes the processes of steps S15 to S18 for the per-domain route for domain B, which is the route that passes through nodes 6, 8, 9, and 11.

In domain B, the label sequence of the path that passes through nodes 6, 8, 9, and 11 is 6, 11. For the path that passes through nodes 11X, 12, and 14 in domain C, the initial label sequence is 11X, 14, and there are two paths, one that passes through nodes 11X, 12, and 14, and the other that passes through nodes 11X, 13, and 14, so the label of node 12 is added in step S18. There is only one minimum-cost path indicated by the label sequence of 11X, 12, and 14, so the label sequence for domain C is 11X, 12, 14.

Returning to FIG. 4, the label calculation process will be described further. In step S19, the label calculation unit 114 combines the label strings of the per-domain routes calculated in steps S14 to S18. When combining, the label calculation unit 114 deletes one of the labels of the boundary nodes. To give a specific example, the label calculation unit 114 combines the label string of 1, 2, 6X, the label string of 6, 11, and the label string of 11X, 12, 14 to calculate the label string of 1, 2, 6X, 11, 12, 14.

In step S20, the label calculation unit 114 outputs the minimum cost route calculated in step S11 and the label strings of each minimum cost route. As a specific example, the label calculation unit 114 outputs the minimum cost route shown in step S11, the label string of 1, 2, 6X, 11, 12, 14, the label string of 1, 4, 6X, 11, 12, 14, the label string of 1, 2, 6X, 11, 13, 14, and the label string of 1, 4, 6X, 11, 13, 14.

<Features of the route calculation device>
The path calculation device 100 calculates a minimum-cost path between a start node and an end node of a path (measurement path) in a network to be measured for delay, jitter, etc. Next, the path calculation device calculates a label string including labels indicating relay nodes on the path. The path calculation device 100 calculates the label string such that the minimum-cost path passing through the calculated label string is only the measurement path, and the number of labels is small. Furthermore, when the network is composed of multiple domains, the path calculation device 100 calculates the label string such that the measurement path passes through the boundary nodes of the domains.

With this type of route calculation device 100, it is possible to measure the delay and jitter of the measurement route using measurement packets with a reduced number of labels. This makes it possible to measure even large-scale networks that could not be measured using the labels of all relay nodes in the past. Furthermore, the route calculation device 100 has a similar effect in multi-domain networks.

<Modification: Label Calculation Process>
In step S18 in FIG. 4, the label calculation unit 114 adds the label of the first node not included in the label string, but it may be the last node not included in the label string.
Also, instead of calculating the label string while adding labels, the label string may be calculated while deleting labels from a label string that includes all of the labels indicating the nodes of the minimum cost route. In other words, the label calculation unit 114 may calculate the label string by repeatedly deleting one label from a label string that includes the labels indicating all of the nodes included in the minimum cost route, as long as the label string after the deletion matches the minimum cost route.

As another method, the process shown in Fig. 5, which will be described later, may be used as a modified example of the process for calculating the label string of a per-domain route (see steps S15 to S18). Fig. 5 is a flowchart of the label calculation process for a per-domain route according to a modified example of this embodiment. In the following, the network 500 in Fig. 2 is regarded as one domain, and a specific example is a route passing through nodes 1, 2, 3, 6, 8, 9, 11, 12, and 14.
In step S31, the label calculation unit 114 calculates a label string in which intermediate nodes between the start node and the end node of the domain-specific route are deleted.

In step S32, the label calculation unit 114 starts the process of repeating steps S33 to S37 for each intermediate node. Hereinafter, the intermediate node to be treated as a target node will be referred to as a target node.
In step S33, the label calculation unit 114 adds the target node to the label string.
In step S34, the label calculation unit 114 calculates a minimum cost path that passes through a node of a label included in the label string.

In step S35, if the number of minimum cost routes this time is less than the number of minimum cost routes calculated in the previous iterative process in step S34 (step S35→YES), the label calculation unit 114 proceeds to step S37. If the number of minimum cost routes has not decreased (step S35→NO), the label calculation unit 114 proceeds to step S36.
In step S36, the label calculation unit 114 deletes the label added in step S33 from the label string.
In step S37, if the number of minimum cost paths calculated in step S34 is one, the label calculation unit 114 ends the process of FIG.

A specific example will be explained below. The label sequence calculated in step S31 is 1, 14. There are four minimum cost routes passing through the nodes included in this label sequence: a route passing through nodes 1, 2, 3, 6, 8, 9, 11, 12, 14; a route passing through nodes 1, 4, 5, 6, 8, 9, 11, 12, 14; a route passing through nodes 1, 2, 3, 6, 8, 9, 11, 13, 14; and a route passing through nodes 1, 4, 5, 6, 8, 9, 11, 13, 14.

In this specific example, steps S33 to S37 are executed while the target node changes to nodes 2, 3, 6, 8, 9, 11, and 12. The label string resulting from step S33 in the first iteration is 1, 2, and 14. There are two minimum-cost routes passing through the nodes included in this label string: a route passing through nodes 1, 2, 3, 6, 8, 9, 11, 12, and 14, and a route passing through nodes 1, 2, 3, 6, 8, 9, 11, 13, and 14. Since this is fewer than the initial four, step S35 returns YES. However, since there is not just one minimum-cost route, the target node is set to node 3 and the process returns to step S33.

In the next iteration, the label sequence becomes 1, 2, 3, 14, but there are still two minimum cost paths, so step S35 returns NO and the label sequence becomes 1, 2, 14. The same is true for nodes 6, 8, 9, and 11, where the label sequence remains 1, 2, 14 and there are still two minimum cost paths.
In the next iteration, the label sequence becomes 1, 2, 12, 14, resulting in one minimum cost route, and this label sequence is output as the label sequence of the per-domain route (when the network 500 is regarded as one domain).

Other variations
Although several embodiments of the present invention have been described above, these embodiments are merely examples and do not limit the technical scope of the present invention. For example, the path calculation device 100 calculates the minimum cost path and its label string between the start node and end node of a path (measurement path) in the network 500 that is the target of measurement of delay, jitter, etc. Not limited to the measurement path, it may be a path of general data (user data).

The present invention may take various other embodiments, and various modifications such as omissions and substitutions may be made without departing from the gist of the present invention. These embodiments and their modifications are included within the scope and gist of the invention described in this specification, etc., and are included in the scope of the invention described in the claims and their equivalents.

<Hardware Configuration>
The path calculation device 100 according to the embodiment described above is realized by a computer 900 having a configuration as shown in Fig. 6, for example. Fig. 6 is a hardware configuration diagram showing an example of the computer 900 that realizes the functions of the path calculation device 100 according to the embodiment described above. The computer 900 includes a CPU 901, a ROM 902, a RAM 903, an SSD 904, an input/output interface 905 (referred to as an input/output I/F (Interface) in Fig. 6), a communication interface 906 (referred to as a communication I/F in Fig. 6), and a media interface 907 (referred to as a media I/F in Fig. 6). The computer 900 may include a hard disk drive (HDD) instead of the SSD 904, or may include a HDD in addition to the SSD 904.

The CPU 901 operates based on a program stored in the ROM 902 or the SSD 904, and performs control by the control unit 110 shown in Fig. 1. The ROM 902 stores a boot program executed by the CPU 901 when the computer 900 is started, programs related to the hardware of the computer 900, and the like.
The CPU 901 controls an input device 910 such as a mouse or a keyboard, and an output device 911 such as a display or a printer, via an input/output interface 905. The CPU 901 acquires data from the input device 910 via the input/output interface 905, and outputs generated data to the output device 911.

The SSD 904 stores programs executed by the CPU 901 and data used by the programs, etc. The communication interface 906 receives data from other devices (not shown) (e.g., nodes of the network 500) via the communication network and outputs the data to the CPU 901, and also transmits data generated by the CPU 901 to other devices via the communication network.
The media interface 907 reads a program or data stored in the recording medium 912 and outputs it to the CPU 901 via the RAM 903. The CPU 901 loads the program from the recording medium 912 onto the RAM 903 via the media interface 907, and executes the loaded program. The recording medium 912 is an optical recording medium such as a DVD (Digital Versatile Disk), a magneto-optical recording medium such as an MO (Magneto Optical disk), a magnetic recording medium, a conductive memory tape medium, a semiconductor memory, or the like.

For example, when the computer 900 functions as the path calculation device 100 according to the embodiment described above, the CPU 901 of the computer 900 executes the program 128 (see FIG. 1) loaded onto the RAM 903, thereby realizing the functions of the path calculation device 100. The CPU 901 reads the program 128 from the recording medium 912 and executes it. Alternatively, the CPU 901 may read the program from another device via a communication network, or may install the program 128 from the recording medium 912 onto the SSD 904 and execute it.

Effect
The effects of the device will be described below.

The path calculation device 100 according to the embodiment described above is a communication network (see network 500) including nodes and links to which costs are assigned, and includes a minimum-cost path calculation unit 113 that calculates a minimum-cost path from a start node to an end node in the communication network, in which a packet to which a label string, which is a string of labels indicating nodes that the packet will pass through during forwarding, is assigned is forwarded through the nodes indicated by the labels included in the label string so as to minimize the cost.
The path calculation device 100 also includes a label calculation unit 114 that identifies one minimum cost path and calculates a label string of the identified minimum cost path with the number of labels smaller than the number of nodes included in the identified minimum cost path.

With this type of route calculation device 100, the minimum cost route from the start node to the end node is calculated. Furthermore, a label string is calculated that has a smaller number of labels than the number of nodes on the minimum cost route, and that causes packets to pass through the minimum cost route. Calculating a label string like this is useful when the scale of the communication network (network 500) is large, and the length of the label string, which includes labels indicating all nodes on the route, exceeds the specifications of the nodes.

The label calculation unit 114 included in the path calculation device 100 according to the embodiment described above calculates a label string by repeatedly adding a label indicating one of the nodes included in the minimum cost route to a label string including a label indicating a start node and a label indicating an end node, until the route along which a packet to which the added label string is assigned is forwarded matches the minimum cost route.

With this type of route calculation device 100, it becomes possible to calculate a label string that results in a single minimum-cost route while adding labels from the minimum-length label string.

The label calculation unit 114 included in the path calculation device 100 according to the embodiment described above calculates the label string by repeatedly deleting one label from a label string that includes labels indicating all nodes included in the minimum-cost route, as long as the route along which a packet to which the deleted label string is assigned is forwarded matches the minimum-cost route.

With this type of route calculation device 100, it becomes possible to calculate a label string that results in a single minimum-cost route while deleting labels from the maximum-length label string.

The label calculation unit 114 provided in the path calculation device 100 according to the embodiment described above adds a label indicating one of the nodes included in the minimum cost path to a label string including a label indicating a start node and a label indicating an end node, and if the number of paths along which a packet with the added label string is forwarded is smaller than the number of paths along which a packet with the label string before the addition is forwarded, performs the addition, and calculates the label string by repeating the addition until the path along which a packet with the added label string is forwarded matches the minimum cost path.

With this type of path calculation device 100, it becomes possible to add labels from a minimum-length label string while avoiding the addition of labels that do not lead to a reduction in the minimum-cost path indicated by the label string, and to calculate a label string that results in a single minimum-cost path.

The communication network related to the path calculation device 100 according to the embodiment described above is made up of a plurality of domains.
When the start node and the end node are in different domains, the minimum cost path calculation unit 113 calculates a minimum cost path that passes through a boundary node on the boundary of the domains.
The label calculation unit 114 calculates a label string for each partial least-cost route, which is the least-cost route in a plurality of domains, and combines the label strings to calculate a label string to be assigned to packets passing through the least-cost route.

With this type of route calculation device 100, it becomes possible to calculate a minimum-cost route that corresponds to a multi-domain network, and to calculate a label sequence that corresponds to the minimum-cost route.

100 Path calculation device 111 Network information collection unit 112 Measurement unit 113 Minimum cost path calculation unit 114 Label calculation unit 128 Program 130 Network information database 140 Measurement result database 500 Network (communication network)

Claims (7)

1. a minimum cost path calculation unit that calculates a minimum cost path from a start node to a destination node in a communication network including nodes and links to which costs are assigned, in which a packet to which a label string is assigned, which is a string of labels indicating nodes that the packet will pass through during transmission, is transmitted so as to minimize cost by passing through nodes indicated by the labels included in the label string;
   a label calculation unit that identifies one of the minimum cost routes and calculates a label string of the identified minimum cost route having a number of labels smaller than the number of nodes included in the identified minimum cost route.
2. The label calculation unit
   2. The path calculation device according to claim 1, wherein the label string is calculated by repeatedly adding a label indicating any one of the nodes included in the minimum cost route to a label string including a label indicating the start node and a label indicating the end node, until a route along which a packet to which the added label string is assigned matches the minimum cost route.
3. The label calculation unit
   2. The path calculation device according to claim 1, wherein the label string is calculated by repeatedly deleting one label from a label string that includes labels indicating all of the nodes included in the minimum cost path, as long as a route along which a packet to which the label string after the deletion is assigned matches the minimum cost path.
4. The label calculation unit
   adding a label indicating any one of the nodes included in the minimum cost route to a label string including a label indicating the start node and a label indicating the end node, if the number of routes along which a packet to which the label string after the addition is assigned is smaller than the number of routes along which a packet to which the label string before the addition is assigned is assigned, the addition is performed;
   The path calculation device according to claim 1 , further comprising: a label string calculation unit that calculates the label string by repeating the addition until a route along which a packet to which the added label string is assigned coincides with the minimum cost route.
5. The communication network includes:
   It consists of multiple domains,
   The minimum cost path calculation unit
   If the start node and the end node are in different domains, calculating the minimum cost path through a boundary node at the boundary of the domain;
   The label calculation unit
   2. The path calculation device according to claim 1, further comprising: calculating a label string for each partial least-cost path that is the least-cost path in a plurality of domains; and combining the label strings to calculate a label string to be assigned to a packet that follows the least-cost path.
6. A program for causing a computer to function as the route calculation device described in claim 1.
7. The route calculation device
   A communication network including nodes and links to which costs are assigned, in which a packet to which a label string is assigned, which is a string of labels indicating nodes that the packet will pass through during transmission, is transmitted through nodes indicated by the labels included in the label string so as to minimize the cost. The communication network includes: a step of calculating a minimum-cost route from a start node to a destination node;
   identifying one of the minimum-cost routes, and calculating a label string of the identified minimum-cost route having a number of labels smaller than the number of nodes included in the identified minimum-cost route.

Images