JP2012044583A - Route server, transfer device, route control method and network system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a route server, a transfer device, a route control method and a network system, capable of switching a route without waiting for route calculation by the route server when a fault occurs in a link between the transfer devices.SOLUTION: The route server 1 calculates, as route information 300, primary route information 301 and backup route information 302 to be used when a fault of a route occurs, and distributes them to the respective transfer devices 2. The respective transfer devices 2 store the primary route information 301 and the backup route information 302 distributed from the route server 1, and when a fault occurs in the route, switch the route information 300 to be referred to from the primary route information 301 to the backup route information 302, and transfer packets.

Description

  The present invention relates to a route server, a transfer device, a route control method, and a network system technology.

  It is important for network operators to implement new network services flexibly and at low cost by appropriately adding and expanding functions of the transfer network. A current general transfer device (router, etc.) has a structure in which a control system function that provides a route calculation function and a data transfer system that actually performs packet transfer are physically integrated in the same housing. Yes. Therefore, there is a problem that it is not possible to flexibly cope with function addition / extension required for the network.

As a method for solving such a problem, an architecture has been proposed in which a control system function and a data transfer system function are physically separated and distributed on a network (see, for example, Non-Patent Document 1).
FIG. 19 is a diagram illustrating an example of a conventional network system, where (a) is an autonomous distributed network system, and (b) is a diagram illustrating a network system based on the architecture described in Non-Patent Document 1.

In the autonomous distributed network system 1901 shown in FIG. 19A, each transfer device 1902 calculates and stores path information.
On the other hand, in the network system 1951 in which the control system and the data system described in Non-Patent Document 1 shown in FIG. 19B are separated, the control system functions such as the route calculation function and network information collection are centrally deployed in the route server 1952. Only the data transfer function is provided in the transfer device 1953. The route server 1952 calculates route information 1954 of all transfer devices 1953 and distributes this route information to each transfer device 1953.

  The transfer device 1953 performs data transfer (packet transfer) according to the route information 1954 received from the route server 1952. In the architecture of FIG. 19B, the addition of functions such as the expansion of the control protocol and the change of the route calculation method affects only the control system functions. However, it may be performed only for the route server 1952. For this reason, according to the network system 1951 of FIG. 19B, a new network service can be realized flexibly and at low cost. Further, in the architecture shown in FIG. 19 (a), it is difficult to add a function at the research stage to the commercial transfer apparatus 1902 for the purpose of evaluation, but an architecture in which the control system and the data system are separated (FIG. 19). In (b)), it is only necessary to add a function to the route server 1952 managed by the carrier.

M. Caesar, D. Caldwell, N. Feamster, J. Rexford, A. Shaikh, and J. V. D. Merwe, “Design and Implementation of a Routing Control Platform,” in proceedings of NSDI 2005.

However, in the architecture described in Non-Patent Document 1, when a failure of a node (transfer device) or a link constituting the network occurs, the route server 1952 collects network information after the failure, performs route calculation, and again It must be distributed to all transfer devices. At this time, the route server 1952 needs to calculate route information and create route information for all the transfer devices, increasing the processing load. For example, when the route calculation function is provided in the transfer device 1902 as in the architecture shown in FIG. 19A, each transfer device 1902 needs to create only its own route information, while the control system and the data system are In the separated architecture of FIG. 19B, since the path server 1952 needs to calculate path information for the node (transfer device 1953), the calculation load increases compared to the architecture of FIG. 19A. For this reason, in the architecture in which the control system and the data system in FIG. 19B are separated, there is a problem that the path cannot be immediately switched with respect to a network state change such as a failure.
Further, the architecture shown in FIG. 19B has a problem that when a failure occurs in the route server 1952, the route cannot be switched.

  That is, in the architecture in which the control system and the data system in FIG. 19B (Non-patent Document 1) are separated (that is, the calculation of route information is concentrated on the route server 1952), the control system function and the data transfer system function are physically separated. By separating them and distributing them on the network, it is possible to provide flexibility and low cost for adding and expanding functions of the control system. On the other hand, since it is necessary for the route server to manage and calculate route information of all the transfer devices, there is a problem that it is difficult to immediately switch the route in response to a network state change such as a failure.

  The present invention has been made in view of such a background, and the present invention is a path server capable of switching a path without waiting for a path calculation by the path server when a failure occurs in a link between transfer apparatuses, It is an object of the present invention to provide a transfer device, a route control method, and a network system.

  The present invention was created to solve the above-described problem, and the path server according to claim 1 collects adjacent information, which is information related to an adjacent relationship between the transfer devices, from the transfer device, and collects the information. A route server that calculates route information that is a transfer route of communication data based on the adjacent information and distributes the calculated route information to each of the transfer devices, and connects the transfer devices as the route information. A path information calculation unit that calculates primary path information used before a path failure occurs and one or more backup path information used when the path failure occurs; the calculated primary path information; and the backup path information And a route setting unit that distributes the information to each of the transfer devices.

  According to this invention, by distributing the backup route information for when a failure occurs to each transfer device, the route server can recalculate the route information even if a failure occurs in the link between the transfer devices. The route can be switched without waiting.

  The transfer device according to claim 2 is a transfer device that transfers received communication data with reference to route information stored in its storage unit, and is a primary route distributed from the route server. Information, the storage unit storing the one or more backup route information, and when a failure occurs in the route, the reference route information is switched from the primary route information to the backup route information, and And a packet control unit for transferring communication data.

  According to this invention, even if a failure occurs in the link between the transfer apparatuses, by switching the reference route information from the primary route information to the backup route information, without waiting for recalculation of the route information by the route server, The route can be switched. Further, even when a failure occurs in the route server and the route information cannot be recalculated when the failure occurs, the backup route information can be referred to, so that the transfer of communication data can be continued.

  Further, the transfer device according to claim 3 is the transfer device according to claim 2, wherein the packet control unit has a failure in the route when the communication data is transferred by the backup route information. If it occurs, the received communication data is discarded.

  According to this invention, it is possible to prevent a loop of communication data that may occur by switching from a state in which backup path information is used to another backup path information.

  The route control method according to claim 4, wherein the route server collects adjacent information, which is information related to the adjacent relationship between the transfer devices, from the transfer device, and transfers communication data based on the collected adjacent information. By a network system that calculates route information that is a route, distributes the calculated route information to each of the transfer devices, and the transfer device transfers received communication data with reference to the distributed route information In the route control method, the route server uses, as the route information, primary route information used before the failure of the route connecting the transfer apparatuses and one or more backup route information used when the route failure occurs. Are calculated and distributed to each of the transfer devices, and each of the transfer devices includes the primary route information distributed from the route server, and the backup route. Information is stored in a storage unit, and when a failure occurs in the route, the reference route information is switched from the primary route information to the backup route information, and the communication data is transferred. .

  According to this invention, even if a failure occurs in the link between the transfer apparatuses, by switching the reference route information from the primary route information to the backup route information, without waiting for recalculation of the route information by the route server, The route can be switched. Further, even when a failure occurs in the route server and the route information cannot be recalculated when the failure occurs, the backup route information can be referred to, so that the transfer of communication data can be continued.

  The route control method according to claim 5 is the route control method according to claim 4, wherein the network system is divided into a plurality of areas, and which route server belongs to which area. Whether to distribute the route information to a device is set, and each of the plurality of route servers distributes the primary route information to the transfer device in the area to which the route information is distributed, and Backup path information is calculated, and the calculated primary path information and backup path information are distributed to transfer devices in the area to which the path information is to be distributed.

  According to this invention, the processing load of each route server can be reduced by distributing the calculation of route information to a plurality of route servers.

  Further, the route control method according to claim 6 is the route control method according to claim 5, wherein the plurality of areas are distributed by the master route server which is one of the plurality of route servers. A first area that is a target, and a second area that a sub-route server that is a route server different from the master route server is a target of distribution of route information, and the sub-route server includes: Collecting neighbor information from transfer devices belonging to the second area, calculating topology information, which is information relating to arrangement of transfer devices in the second area, based on the collected neighbor information, and calculating the topology information To the master route server, and the master route server collects adjacent information from the transfer devices belonging to the first area, and from the sub route server, All areas obtained by collecting topology information of the second area and combining the first area and the second area based on the collected neighboring information of the first area and the topology information of the second area. The topology information of the entire area is distributed to the sub route server, and each of the master route server and the sub route server is configured based on the topology information of the entire area. The path information of the transfer device belonging to the area to be distributed is calculated.

  According to this invention, since the collection of adjacent information in all areas does not concentrate on one route server, the communication load when collecting adjacent information can be reduced. Further, since the information transmitted / received between the route servers is topology information, the communication load between the route servers can be reduced.

  The route control method according to claim 7 is the route control method according to claim 6, wherein the master route server stores information on a transfer device in which a failure occurs between transfer devices in the first area. When the failure information is received, all areas including the first area and the sub-area based on the topology information of the second area stored in its own storage unit and the received failure information The topology information of the entire area is distributed to the sub-route server, and the master route server and the sub-route server are distributed on the basis of the topology information of the entire area. The primary route information of the transfer device belonging to the area is calculated.

  The route control method according to claim 8 is the route control method according to claim 6, wherein the sub-route server has a failure between transfer devices in the second area to which the route information is distributed. When the failure information including the connection information between the generated transfer devices is received, the information including the failure information is transmitted to the master route server, and the master route server stores the second area stored in its own storage unit. The topology information of all areas including the first area and the sub area is calculated on the basis of the topology information and the transmitted failure information, and the topology information of the all areas is calculated to the sub route server. The master route server and the sub route server are based on the topology information of all the areas, and the transfer route belonging to the area to which the route information is to be distributed. And calculating a primary route information of the device.

  According to the seventh and eighth aspects of the present invention, even if a failure occurs in each area, a new primary route can be calculated and distributed while maintaining the effect of the sixth aspect.

  The network system according to claim 9, wherein the route server collects adjacency information that is information related to adjacency relationships between the transfer devices from the transfer device, and is a communication data transfer path based on the collected adjacency information. A network system that calculates route information, distributes the calculated route information to each of the transfer devices, and the transfer device refers to the route information distributed and transfers received communication data. The route server calculates, as the route information, primary route information used before failure of a route connecting the transfer apparatuses and one or more backup route information used when the failure of the route occurs, Each transfer device distributes to each transfer device, and each transfer device stores the primary route information distributed from the route server and the backup route information. Storing, when a failure in the path occurs, the path information to reference, from the primary routing information, switching to the backup path information, and performs the transfer of the communication data.

  According to this invention, the processing load of each route server can be reduced by distributing the calculation of route information to a plurality of route servers.

  The network system according to claim 10 is the network system according to claim 9, wherein the network system is divided into a plurality of areas, and which route server is transferred to which transfer device belongs to which area. Whether to distribute the route information is set, and each of the route servers calculates the primary route information and the backup route information for the transfer device in the area, and calculates the calculated primary route information and the backup route. The information is distributed to a transfer device in the area that is the distribution target of the route information.

  According to this invention, the processing load of each route server can be reduced by distributing the calculation of route information to a plurality of route servers.

  According to the present invention, it is possible to provide a path server, a transfer apparatus, a path control method, and a network system capable of switching paths without waiting for path calculation by the path server when a failure occurs in a link between transfer apparatuses. Can do.

1 is a diagram illustrating an example of a network system according to a first embodiment. It is a figure which shows the structure of the path | route server which concerns on 1st Embodiment. It is a figure which shows the structural example of the transfer apparatus which concerns on 1st Embodiment. The contents regarding various information used in the first embodiment are shown in a table. It is a figure which shows the example of the topology information which concerns on 1st Embodiment. It is a figure which shows the example of the route information which concerns on 1st Embodiment. It is a figure which shows the outline | summary of routing information delivery. It is a figure which shows the outline | summary of the path | route switching which concerns on 1st Embodiment (the 1). It is a figure which shows the outline | summary of the path switching which concerns on 1st Embodiment (the 2). It is a flowchart which shows the flow of the calculation process of the route information in the normal time which concerns on 1st Embodiment. It is a flowchart which shows the flow of the calculation process of the route information at the time of the failure generation which concerns on 1st Embodiment. It is a flowchart which shows the flow of the path | route information reception and setting process in the transfer apparatus which concerns on 1st Embodiment. It is a flowchart which shows the flow of the transfer process which concerns on 1st Embodiment. It is a figure which shows the structural example of the network system which concerns on 2nd Embodiment. It is a flowchart which shows the flow of the path | route information calculation process (normal time) in the master path | route server which concerns on 2nd Embodiment. It is a flowchart which shows the flow of the route information calculation process (normal time) in the sub route server which concerns on 2nd Embodiment. It is a flowchart which shows the flow of the route information calculation process (at the time of failure occurrence) in the master route server concerning 2nd Embodiment. It is a flowchart which shows the flow of the route information calculation process (at the time of failure occurrence) in the sub route server concerning 2nd Embodiment. It is a figure which shows the structural example of the conventional network system.

  Next, modes for carrying out the present invention (referred to as “embodiments”) will be described in detail with reference to the drawings as appropriate.

[First Embodiment]
First, a first embodiment of the present invention will be described with reference to FIGS.
(Network system)
FIG. 1 is a diagram illustrating an example of a network system according to the first embodiment.
The network system 10 includes a transfer device 2 such as a router, and a route server 1 that calculates route information 300 for these transfer devices 2 and distributes the calculated route information 300 to each transfer device 2.
In addition to the primary route calculation unit (route information calculation unit) 102 that calculates the route information 300 (referred to as primary route information 301) that serves as a reference (used before the occurrence of a route failure), the route server 1 uses a network. A backup path calculation unit (path information calculation unit) 103 that calculates spare path information 300 (referred to as backup path information 302) to be used when a failure occurs in each transfer device 2, and the calculated path information 300 (Hereinafter, when described as the route information 300, it indicates both the primary route information 301 and the backup route information 302) is distributed to the corresponding transfer device 2. Note that broken line arrows in FIG. 1 indicate distribution of route information 300.

The connection between the transfer apparatuses 2 is called a link. Traffic is generated between the transfer apparatuses 2, and this traffic arrives at the transfer apparatus 2 as a transfer destination via a link in units of IP (Internet Protocol) packets (hereinafter referred to as packets) which are communication data. Shall.
In FIG. 1, there are a plurality of backup path information 302 (two in FIG. 1) because the path server 1 calculates the backup path information 302 corresponding to each failure pattern for each transfer device 2. This indicates that there are a plurality of backup path information 302 corresponding to and stored in one transfer apparatus 2.

(Route server configuration)
FIG. 2 is a diagram illustrating a configuration of the route server according to the first embodiment. Reference is made to FIG. 1 as appropriate.
The path server 1 includes a processing unit 100, a topology information DB (Data Base) 110, a primary path information DB 120, and a backup path information DB 130 that are storage units.
The processing unit 100 includes a topology information processing unit 101, a primary route calculation unit 102, a backup route calculation unit 103, a route setting unit 104, a failure / return information receiving unit 105, and an operator I / F (Interface) unit 106. Yes.
The topology information processing unit 101 collects adjacent information transmitted from each transfer device 2, calculates the topology information that is the placement information of the transfer device 2 in the network system 10 based on the collected adjacent information, and calculates The topology information is stored in the topology information DB 110.
When the primary route calculation unit 102 calculates the primary route information 301 corresponding to each transfer device 2 by calculating the transfer route based on the calculated topology information, the primary route information 301 is stored in the primary route information DB 120. Based on the calculated topology information, the backup path calculation unit 103 calculates backup path information 302, which is backup path information that is a detour path provided for every failure pattern, for each transfer device 2, and this backup path information. Store in DB130.
The route setting unit 104 transmits the calculated route information 300 (primary route information 301 and backup route information 302) to the corresponding transfer device 2.
The operator I / F unit 106 is an interface with an administrator, and inputs data to a terminal such as a PC (Personal Computer) that uses a server maintenance or backup route information 302 calculation opportunity.

  The topology information stored in the topology information DB 110, the primary path information 301 stored in the primary path information DB 120, and the backup path information 302 stored in the backup path information DB 130 will be described later.

  The path server 1 is realized by a PC or the like, and the processing unit 100 and the units 101 to 106 in the processing unit 100 are stored in a ROM (Read Only Memory) or a HDD (Hard Disk Drive) (not shown). The program is embodied by being expanded in a random access memory (RAM) and executed by a central processing unit (CPU).

(Transfer device)
FIG. 3 is a diagram illustrating a configuration example of the transfer device according to the first embodiment. Reference is made to FIG. 1 as appropriate.
The transfer device 2 includes an operation control unit 200, a packet control unit 210, a primary route information DB (storage unit) 220, and a backup route information DB (storage unit) 230.
The operation control unit 200 performs settings related to the route information 300, and includes an adjacent information transmission unit 201, a failure / recovery information transmission unit 202, a route information reception unit 203, and a hardware management unit 204. The packet control unit 210 performs processing related to transfer of the packet 400 and the like, and includes a state change detection unit 211 and a packet transfer unit 212.

The adjacency information transmission unit 201 transmits adjacency information that is information on the adjacency relationship with the transfer apparatus 2 connected to itself.
The failure / return information transmission unit 202 transmits failure information, which is information on a failure (link failure) detected by the state change detection unit 211, and return information, which is return information, to the path server 1. The failure information and the return information are, for example, information on a set of a node ID (Identification) of itself and a node ID of the partner transfer apparatus 2 that is disconnected.
The route information receiving unit 203 receives the route information 300 (primary route information 301 and backup route information 302) distributed from the route server 1.
The hardware management unit 204 stores the distributed route information 300 in the primary route information DB 220 and the backup route information DB 230. The hardware management unit 204 also has a maintenance function for performing resetting of itself.

The state change detection unit 211 detects a link failure or link recovery, and generates failure information and recovery information.
The packet transfer unit 212 transfers the packet 400 to the next transfer device 2 by referring to the route information 300. When a failure is detected, switching to appropriate backup path information 302 is performed.

The primary route information 301 stored in the primary route information DB 220 and the backup route information 302 stored in the backup route information DB 230 will be described later.
Also, as shown in FIG. 3, the packet 400 has a route ID, which will be described later, in the header portion.

  The operation control unit 200, the packet control unit 210, and each of the units 201 to 204, 211, and 212 in the transfer device 2 are realized by a program stored in a ROM or the like being executed by the CPU.

(Various information)
FIG. 4 is a table showing contents related to various information used in the first embodiment.
The topology information has a connection relationship between nodes and links as information elements, and is information that the topology information processing unit 101 of the route server 1 collects from each transfer device 2 and is stored in the topology information DB 110.
The primary route information 301 is route information 300 used when a failure is not detected, and includes a route ID, a packet destination address, and an output interface (I / F) of the transfer apparatus 2 as information elements. . Here, the output interface may be a next hop address. The primary route information 301 is information that the primary route calculation unit 102 of the route server 1 calculates for each transfer device 2 based on the topology information and distributes it to the corresponding transfer device 2. The route ID is an ID for uniquely identifying the route, and “0” is set in the primary route information 301.

  The backup path information 302 is path information 300 that is used after a failure occurs, and includes a path ID, a destination address, and an output interface (I / F) as information elements. The output interface may be a next hop address. As described above, the backup path information 302 is information that is calculated by the backup path calculation unit 103 of the path server 1 for each failed transfer apparatus 2 based on the topology information and distributed to the corresponding transfer apparatus 2. is there. The path ID of the backup path information 302 stores a value other than “0”.

  Note that the primary route information DB 220 and the backup route information DB 230 of the transfer device 2 may be CAM (Content-Addressable Memory) or the like in order to speed up hardware transfer.

(Topology information)
FIG. 5 is a diagram illustrating an example of topology information according to the first embodiment.
The topology information includes a start node ID, an end node ID, a connection state, a link cost, and the like. The link cost is a value defined by the routing protocol used. Note that the node is the transfer device 2, and hereinafter, the transfer device 2 may be appropriately referred to as a node.
In other words, the topology information is information describing the adjacency relationship in all the transfer devices 2, and “None” is registered in the connection column between the transfer devices 2 not having the connection relationship.

(Route information)
FIG. 6 is a diagram illustrating an example of route information according to the first embodiment. The configuration of the route information 300 shown here is common to the primary route information 301 and the backup route information 302.
The route information 300 includes a route ID for distinguishing the route information 300, a destination address, and an output interface (I / F). As described above, the output interface (I / F) may be a next hop address.

(Overview)
Here, an overview of the first embodiment will be described with reference to FIGS.
FIG. 7 is a diagram showing an outline of route information distribution. FIG. 7A is a schematic diagram of distribution of route information 300 by the technique (comparative example) described in Non-Patent Document 1, and FIG. 7B is a schematic diagram of distribution of route information 300 according to the present embodiment.
As shown in FIG. 7A, in the technique described in Non-Patent Document 1, one route information 300 is distributed to one transfer device 2, whereas the present embodiment shown in FIG. In the embodiment, a plurality of route information 300 such as primary route information 301 and backup route information 302 is distributed to one transfer device 2. Each route information 300 (route) is distinguished by a route ID. In other words, each route information 300 is encapsulated by the route ID.

8 and 9 are diagrams showing an outline of the path switching according to the first embodiment.
FIG. 8 shows a case where the packet 400 (FIG. 3) is transmitted from the transfer device 2a to the transfer device 2c. As shown in FIG. 3, the packet 400 describes a route ID to be used.
As shown in FIG. 8A, in the primary route (route ID “0”), the packet 400 is transmitted in the order of the transfer device 2a → the transfer device 2b → the transfer device 2c (arrow 801), but the transfer device 2b, When the failure of the link between 2c is detected, the packet 400 transmitted from the transfer device 2a stops at the transfer device 2b (arrow 802). When the transfer device 2b selects an appropriate one from the backup route information 302 (assuming that the route ID “1” is selected), the transfer device 2b rewrites the route ID in the packet 400 from “0” to “1” and is selected. The rewritten packet 400 is transmitted to the new transfer apparatus 2 according to the backup path information 302. In the case of FIG. 8A, the transfer device 2b returns the packet 400 to the transfer device 2a (arrow 803).

  When the transfer device 2a that has received the packet 400 with the rewritten route ID acquires the backup route information 302 (route ID “1”) from the backup route information DB 230 using the route ID described in the packet 400 as a key, The packet 400 is transmitted according to the backup path information 302 (transfer device 2a → transfer device 2d → transfer device 2e → transfer device 2f → transfer device 2c: arrow 804).

  The transfer device 2b sends a failure information to the effect that a failure has occurred in the link between the transfer devices 2b and 2c to the path server 1 while detecting the disconnection of the link. The failure information is information that the transfer device 2 sends to the path server 1 when the existence information is not sent from the confirmation partner in the periodic existence check in the transfer device 2, and the node ID of the own transfer device 2 , Information that is paired with the node ID of the confirmation destination. In this way, the failure information describes the node ID of its own transfer device 2 and the node ID of the link destination transfer device 2 in the same manner as the adjacent information. The primary route information 301 is recalculated. Then, the route server 1 distributes the recalculated primary route information 301 (transfer device 2a → transfer device 2d → transfer device 2e → transfer device 2f → transfer device 2c) to each transfer device 2. Each transfer device 2 updates the primary route information 301 stored in its own primary route information DB 220 to the newly received primary route information 301, and transfers the packet 400 according to the updated primary route information 301 (broken arrows). 805).

  As shown in FIG. 8B, when each transfer device 2 transfers the packet 400 according to the recalculated primary route information 301 (arrow 805), the return of the link between the transfer devices 2b and 2c is detected. Then, the transfer device 2b transmits return information to the effect that the link between the transfer devices 2b and 2c has been returned to the route server 1. The return information is information that the transfer device 2 sends to the route server 1 when the survival information is sent from a confirmation partner that has not been confirmed to be alive in the periodic life check in the transfer device 2. This is information in which the node ID of the own transfer device 2 and the node ID of the confirmation destination are paired. As described above, since the return information includes its own node ID and link destination node ID in the same manner as the adjacent information, the route server 1 that has received the return information uses the primary route information based on the return information. 301 is calculated again. Then, the route server 1 distributes the recalculated primary route information 301 (transfer device 2a → transfer device 2b → transfer device 2c) to each transfer device 2, and each transfer device 2 follows the newly received primary route information 301. The packet 400 is transferred (broken arrow 806). Note that the transfer device 2 does not switch to the backup path information 302 even if a return is detected. This is because the route switching at the time of return is not urgent.

FIG. 9 is a diagram showing an outline of route switching according to the first embodiment.
FIG. 9A is a diagram showing an example of primary route information 301 (route ID “0”), in which a route for transmitting the packet 400 is described in the order of transfer device 2A → transfer device 2B → transfer device 2C. It shows that.
FIG. 9B is a diagram showing an example of the backup path information 302 (path ID “1”). In this backup path, as a path when a failure occurs in the link between the transfer apparatuses 2A and 2B, The path of transfer device 2A → transfer device 2D → transfer device 2B → transfer device 2C is described.
FIG. 9C is a diagram showing an example of another backup route information 302 (route ID “2”). In this backup route, as a route when the link between the transfer apparatuses 2B and 2C is shut off, The path of transfer device 2A → transfer device 2D → transfer device 2E → transfer device 2C is described. In FIG. 9C, after the packet 400 is transmitted to the transfer apparatus 2B, the packet 400 returns to the transfer apparatus 2A. As described in FIG. 8, the packet 400 sent to the transfer apparatus 2B is The route ID is rewritten from “0” to “2” by the transfer device 2B, and the packet 400 is transmitted to the new transfer device 2 according to the rewritten backup route information 302. As a result, the route ID is returned to the transfer device 2A. ing.

<Processing>
Next, processing according to the present embodiment will be described with reference to FIGS. 10 to 13 with reference to FIGS. 2 and 3.
(Route information calculation process (normal time))
FIG. 10 is a flowchart showing the flow of the route information calculation process in the normal time according to the first embodiment. Here, the normal processing means that the administrator inputs information to the route server 1 via the operator I / F unit 106 at the time of starting the network and when the network state changes such as removal / addition of the transfer device 2. This is the process performed by the route server 1.
First, the topology information processing unit 101 collects adjacent information from the adjacent information transmitting unit 201 of each transfer device 2 (S101).
Then, the topology information processing unit 101 calculates topology information based on the collected adjacent information (S102).
Next, the primary path calculation unit 102 calculates primary path information 301 corresponding to each transfer apparatus 2 for each transfer apparatus 2 based on the calculated topology information (S103).
Further, the backup path calculation unit 103 calculates the backup path information 302 for each transfer device 2 based on the acquired topology information (S104). As described above, the backup path information 302 is calculated so that data can be sent to all the transfer apparatuses 2 regardless of which link is disconnected at one place. Therefore, a plurality of backup path information 302 are generated for one transfer apparatus 2.
Then, the route setting unit 104 distributes the calculated route information 300 (primary route information 301 and backup route information 302) to the corresponding transfer devices 2 (S105).

(Route information calculation processing (when a failure occurs))
FIG. 11 is a flowchart showing a flow of a route information calculation process when a failure occurs according to the first embodiment. This process is a process performed by the route server 1.
First, the failure / recovery information receiving unit 105 determines whether failure information has been received from the transfer device 2 (S201).
As a result of step S201, when failure information has not been received (S201 → No), the failure / return information receiving unit 105 advances the processing to step S205.

If failure information is received as a result of step S201 (S201 → Yes), the topology information processing unit 101 receives the failure information because the failure information includes the adjacent information of the link in which the failure has occurred as described above. The topology information is updated based on the failure information (S202). Specifically, the connection information of the two node IDs included in the failure information is changed from “Yes” to “No”.
When the primary route calculation unit 102 recalculates new primary route information 301 for each transfer device 2 according to the updated topology information (S203), the route setting unit 104 sends the new primary route information 301 to each transfer device 2. Distribute (S204).

Next, the failure / return information receiving unit 105 determines whether or not the return information is received from the transfer device 2 (S205).
When the return information is not received as a result of step S205 (S205 → No), the failure / return information receiving unit 105 returns the process to step S201.

When failure information is received as a result of step S205 (S205 → Yes), the return information includes the adjacent information of the returned link, so the topology information processing unit 101 uses the received return information as a basis. The topology information is updated (S206). Specifically, the connection information of the two node IDs included in the return information is changed from “none” to “present”.
When the primary route calculation unit 102 recalculates the new primary route information 301 for each transfer device 2 according to the updated topology information (S207), the route setting unit 104 sends the new primary route information 301 to each transfer device 2. Distribute (S208).

(Route information reception / setting)
FIG. 12 is a flowchart showing a flow of route information reception / setting processing in the transfer apparatus according to the first embodiment.
First, the route information receiving unit 203 receives the route information 300 (primary route information 301 and backup route information 302) distributed in step S105 of FIG. 10, and the route information 300 received by the hardware management unit 204 is changed to the primary route. The route information 300 is set by storing the information in the information DB 220 and the backup route information DB 230 (S301).
The state change detection unit 211 of the transfer device 2 periodically transmits the survival confirmation information to the adjacent transfer device 2 and monitors the link state. That is, the state change detection unit 211 determines whether or not a link failure is detected by not receiving a survival confirmation response from the adjacent transfer device 2 (S302).
If no failure is detected as a result of step S302 (S302 → No), the state change detection unit 211 returns the process to step S302.

If a failure is detected as a result of step S302 (S302 → Yes), the failure / return information transmission unit 202 transmits the node ID of the own transfer device 2 and the destination transfer device 2 that has not received a survival confirmation response. The failure information having the paired node ID and the node information is transmitted to the route server 1 (S303).
Then, the route information receiving unit 203 monitors whether or not the primary route information 301 distributed in step S204 of FIG. 11 has been received from the route server 1 (S304).
If the primary route information 301 has not been received as a result of step S304 (S304 → No), the route information receiving unit 203 returns the process to step S304. If the packet 400 is received before the primary route information 301 from the route server 1 has been received, the processing of “No” is performed in step S404 of FIG. It will be switched to the route by.

  When the primary route information 301 is received as a result of step S304 (S304 → Yes), the hardware management unit 204 updates the primary route information 301 held in the primary route information DB 20 to the received primary route information 301. (S305). Thereafter, the packet 400 is transferred according to the primary route information 301 updated in step S305. That is, the backup route information 302 is the emergency route information 300 for the time when the primary route information 301 is being recalculated.

Then, the state change detection unit 211 determines whether or not link recovery is detected by receiving a survival confirmation response from the adjacent transfer device 2 (S306).
If no return is detected as a result of step S306 (S306 → No), the state change detection unit 211 returns the process to step S306.

When the return is detected as a result of step S306 (S306 → Yes), the failure / return information transmission unit 202 receives the node ID of the own transfer device 2 and the transfer device 2 of the other party that has obtained the survival confirmation response. The return information having the node ID and the pair information is transmitted to the route server 1 (S307).
Then, the route information receiving unit 203 monitors whether or not the primary route information 301 distributed in step S208 of FIG. 11 has been received from the route server 1 (S308).
If the primary route information 301 has not been received as a result of step S308 (S308 → No), the route information receiving unit 203 returns the process to step S308.

  If the primary route information 301 is received as a result of step S308 (S308 → Yes), the hardware management unit 204 updates the primary route information 301 held in the primary route information DB 220 to the received primary route information 301. (S309).

(Transfer process)
FIG. 13 is a flowchart showing a flow of transfer processing according to the first embodiment. This process is a process performed by each transfer device 2.
The packet transfer unit 212 determines whether the packet 400 has been received (S401).
As a result of step S401, when the packet 400 has not been received (S401 → No), the packet transfer unit 212 returns the process to step S401.
When the packet 400 is received as a result of step S401 (S401 → Yes), the packet transfer unit 212 acquires the route ID described in the header of the packet 400 (S402), and the route information 300 (corresponding to the acquired route ID) ( The primary route information 301 or the backup route information 302) is acquired from the primary route information DB 220 or the backup route information DB 230 (S403).

Next, the packet transfer unit 212 determines whether or not the packet 400 can be transferred by determining whether or not a failure has occurred in the output interface for transferring to the destination address of the packet 400 in the acquired route information 300. Is determined (S404).
If the packet 400 is transferable as a result of step S404 (S404 → Yes), the packet transfer unit 212 transfers the packet 400 via the output interface described in the acquired path information 300 (S405), The process returns to step S401.
If it is determined in step S404 that the packet 400 cannot be transferred (S404 → No), the packet transfer unit 212 determines whether or not the route ID described in the header of the packet 400 is “0” (that is, the current It is determined whether or not the used route information 300 is the primary route information 301 (S406).
As a result of step S406, when the route ID is not “0” (SS406 → No), that is, when the currently used route information 300 is already the backup route information 302, the packet transfer unit 212 determines that the packet 400 is Discard (S407), the packet transfer unit 212 returns the process to step S401. This is to prevent a transfer loop that may occur when the path information 300 is changed to another backup path information 302 while the backup path information 302 is already used.

If the route ID is “0” as a result of step S406 (S406 → Yes), the packet transfer unit 212 selects appropriate backup route information 302 (S408). At this time, since the backup path information 302 describes information such as which backup path information 302 is used when a failure occurs in which link, the packet transfer unit 212 uses the information to back up the backup path information. 302 is selected.
Then, the packet transfer unit 212 rewrites the route ID described in the header of the packet 400 with the route ID of the backup route information 302 selected in step S408 (S409), and the packet 400 is changed to a new transfer destination according to the route ID. Is transmitted to the transfer device 2 (S410). The transfer device 2 (packet transmission source) that has received the returned packet 400 transfers the packet 400 in accordance with the processing from step S401.

Further, when the packet 400 is transferred in accordance with the primary route information 301 before the return (recalculated primary route information 301), if a failure further occurs, the route server 1 displays a warning and notifies the administrator. Then, the calculation processing of the route information 300 in FIG. 15 may be performed according to an instruction from the administrator.
The route information 300 may be distributed using any protocol, for example, ForCES (for example, A. Doria, R. Haas, JH Salim, H. Khosravi, and WM Wang, “ ForCES Protocol Specification, “See IETF Internet draft draft-ietf-forces-protocol-22, Mar. 2009.)

(Summary of the first embodiment)
By setting the backup path information 302 to be used when a link failure occurs in each transfer device 2, the path can be switched without waiting for the route server 1 to recalculate the route information 300 when a link failure occurs.
Further, even if a failure occurs in the route server 1 and the recalculation of the route information 300 becomes impossible, the transfer of the packet 400 can be continued by the backup route information 302.
Furthermore, if a link failure occurs during transfer of the packet 400 according to the backup path information 302, the packet 400 is discarded (communication is stopped) so that the backup path information 302 is used. Thus, it is possible to prevent a routing loop that may occur by switching to the other backup path information 302.
Further, since the route information 300 recalculated when a failure occurs or when it is restored is only the primary route information 301, the calculation load of the route information 300 in the route server 1 or the distribution of the route information 300 to the transfer device 2 The communication load in the network is not increased unnecessarily.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIGS.
In the second embodiment, the effect of the first embodiment is maintained by dividing the network system 10a into a plurality of areas and preparing a plurality of route servers 1 each of which is a distribution target of the route information 300 (FIG. 1). However, the calculation processing load of the route information 300 is reduced.

(System configuration)
FIG. 14 is a diagram illustrating a configuration example of a network system according to the second embodiment.
Here, the network system 10a is divided into a backbone area (first area) 1411 and a sub-area (second area) 1412, and the sub-area 1412 is further divided into a sub-area 1412a and a sub-area 1412b. Any number of subareas 1412 may be set, but only one master path server 1411 is set in the network system 10a. A route server 1 (master route server 1401, sub route servers 1402a and 1402b) that is targeted for distribution in the transfer device 2 in each area is installed. In FIG. 14, the master route server 1401 has a backbone area 1411 as a distribution target, the sub route server 1402a has a sub area 1412a as a distribution target, and the sub route server 1402b has a sub area 1412b as a distribution target.

The sub route server 1402a collects adjacent information of the transfer device 2 in the sub area 1412a (hereinafter referred to as adjacent information of the sub area 1412a. The same applies to the sub area 1412b and the backbone area 1411), and based on the adjacent information. The calculated topology information is transmitted to the master route server 1401. Similarly, the sub route server 1402b collects adjacent information of the sub area b, and transmits the topology information calculated based on the adjacent information to the master route server 1401.
Note that the information transmitted from the sub route server 1402 to the master route server 1401 is not limited to the topology information. If the topology information has already been received once, the adjacent information acquired from each transfer device 2 that is difference information is used. Also good.
The master route server 1401 that has collected the topology information of the sub-areas 1412a and 1412b from the sub-route servers 1402a and 1402b is based on the neighboring information of the backbone area 1411 collected by itself and the topology information of the sub-areas 1412a and 1412b. The topology information of the areas (backbone area 1411, subareas 1412a, 1412b) is calculated, and the calculated topology information of all areas is transmitted to the sub route servers 1402a, 1402b.

Each route server 1 that has topology information of all areas calculates route information 300 (primary route information 301 and backup route information 302) for each area to be distributed.
That is, the master route server 1401 calculates the route information 300 corresponding to the transfer devices 2o, 2p, and 2q, and the sub route server 1402a calculates the route information 300 corresponding to the transfer devices 2r, 2s, and 2t, and The server 1402b calculates route information 300 corresponding to the transfer devices 2u, 2v, and 2w, and distributes the route information 300 to each transfer device 2.

  The configuration of the master route server 1401 and the sub route server 1402 is the same as the configuration of the route server 1 shown in FIG. 2, and the configuration of the transfer device 2 is the same as the configuration of the transfer device 2 shown in FIG. Then, explanation is omitted.

Hereinafter, the operation of each route server 1 in the second embodiment will be described with reference to FIGS. 15 to 18 with reference to FIGS. Since the operation of the transfer device 2 is the same as that of the first embodiment, description and illustration are omitted.
(Route information calculation processing (normal: master route server))
FIG. 15 is a flowchart showing a flow of route information calculation processing (normal time) in the master route server according to the second embodiment. The normal processing in FIG. 15 and FIG. 16 means that when the network state changes such as when the network is started and when the transfer device 2 is removed or added, the administrator inputs information to the route server 1. This is processing performed by the master route server 1401 and each sub route server 1402.
First, the topology information processing unit 101 of the master route server 1401 collects adjacent information from the adjacent information transmitting unit 201 of the transfer device 2 in the backburn area (S501).
Next, the topology information processing unit 101 of the master route server 1401 collects the topology information of each subarea 1402 from each subroute server 1402 (S502).
Then, the topology information processing unit 101 calculates the topology information of all areas from the adjacent information of the backbone area 1411 collected in step S501 and the topology information of each subarea 1412 collected in step S502 (S503). Specifically, when topology information processing unit 101 calculates the topology information of backbone area 1411 based on the adjacent information of backbone area 1411, the topology information of backbone area 1411 and the topology information of each sub-area 1412 are combined. . Connection information between transfer devices 2 (transfer device 2p-transfer device 2r, etc. in FIG. 14) existing at the boundary of each area is described in the topology information of each area.

  Then, the topology information processing unit 101 distributes the calculated topology information of all areas to each sub route server 1402 (S504). At this time, the topology information transmitted to the sub-route server 1402 may be all topology information including the backbone area 1411 and each sub-area 1412, or only necessary topology information may be distributed. For example, since it is meaningless to distribute the topology information of the subarea 1412a to the subroute server 1402a in FIG. 14, only the topology information of the backbone area 1411 and the subarea 1412b may be transmitted to the subroute server 1402a. Good.

Next, the primary route calculation unit 102 and the backup route calculation unit 103 of the master route server 1401 transfer data in the backbone area 1411 based on the calculated topology information of all areas (transfer devices 2o to 2q in FIG. 14). The primary route information 301 and the backup route information 302 (route information 300) are calculated (S505).
Next, the route setting unit 104 of the master route server 1401 distributes the calculated primary route information 301 to the transfer device 2 in the backbone area 1411 (S506).

(Route information calculation processing (sub route server))
FIG. 16 is a flowchart showing a flow of route information calculation processing (normal time) in the sub route server according to the second embodiment.
First, the topology information processing unit 101 of the sub route server 1402 collects adjacent information of the sub area 1412 from the adjacent information transmitting unit 201 of the transfer apparatus 2 belonging to the sub area 1412 to which the route information 300 is distributed (S601). ).
Next, the topology information processing unit 101 calculates topology information in the sub-area 1412 to which the route information 300 is distributed from the collected adjacent information (S602).
Subsequently, the topology information processing unit 101 transmits the calculated topology information of the subarea 1412 to the master route server 1401 (S603).

  The topology information processing unit 101 receives the topology information of all areas distributed from the master route server 1401 in step S504 in FIG. 15 (S604). As described above, the topology information received at this time may be the topology information of all of the backbone area 1411 and each sub-area 1412 or only the necessary topology information may be received.

Then, the primary route calculation unit 102 and the backup route calculation unit 103 of the sub route server 1402 use the received topology information of all areas to transfer devices 2 in the sub area 1412 (transfer devices 2r to 2t in FIG. 14 or Primary path information 301 and backup path information 302 (path information 300) for the transfer apparatuses 2u to 2w) are calculated (S605).
Next, the route setting unit 104 of the sub route server 1402 distributes the calculated primary route information 301 to the transfer device 2 in the sub area 1412 to which the route information 300 is distributed (S606).

(Route information calculation processing (when a failure occurs: master route server))
FIG. 17 is a flowchart showing a flow of route information calculation processing (when a failure occurs) in the master route server according to the second embodiment.
First, the failure / return information receiving unit 105 or the topology information processing unit 101 of the master route server 1401 has received failure information from any transfer device 2 in the backbone area 1411 corresponding to itself, or from the sub route server 1402. It is determined whether topology information has been received (S701).
As a result of step S701, when failure information and topology information are not received (S701 → No), the failure / return information receiving unit 105 and the topology information processing unit 101 advance the processing to step S706.
When failure information or topology information is received as a result of step S701 (S701 → Yes), the topology information processing unit 101 obtains topology information for all areas based on the received failure information or topology information received from the sub-route server 1402. Calculate (S702).
Then, the topology information processing unit 101 distributes the calculated topology information of all areas to each sub route server 1402 (S703).
Then, the primary route calculation unit 102 recalculates the primary route information 301 in the backbone area 1411 based on the topology information of all areas (S704).
Subsequently, the route setting unit 104 distributes the recalculated primary route information 301 to the transfer device 2 in the backbone area 1411 (S705).

Next, when the failure / return information receiving unit 105 of the master route server 1401 receives failure information from the backbone area 1411 in step S701, it determines whether or not the return information has been received. In step S701, the sub route server 1402 is determined. If the topology information is received from the sub route server 1402, the topology information processing unit 101 of the master route server 1401 determines whether or not the topology information is sent again from the sub route server 1402 (S706).
If the return information and the topology information are not received as a result of step S706 (S706 → No), the failure / return information receiving unit 105 returns the process to step S701.

As a result of step S706, when return information or topology information is received (S706 → Yes), the topology information processing unit 101 calculates topology information for all areas based on the received failure information or topology information (S707). . The method for calculating the topology information for all areas is the same as that in step S503 in FIG.
Then, the topology information processing unit 101 distributes the calculated topology information of all areas to each sub route server 1402 (S708).
Then, the primary route calculation unit 102 recalculates the primary route information 301 in the backbone area 1411 based on the topology information of all areas (S709).
Subsequently, the route setting unit 104 distributes the recalculated primary route information 301 to the transfer device 2 in the backbone area 1411 (S710).

(Route information calculation processing (when a failure occurs: sub route server))
FIG. 18 is a flowchart showing a flow of route information calculation processing (when a failure occurs) in the sub route server according to the second embodiment.
First, it is determined whether the failure / return information receiving unit 105 of the sub-route server 1402 has received failure information from any of the transfer devices 2 in the sub-area 1412 corresponding to the sub-route server 1402 (S801).
As a result of step S801, when failure information has not been received (S801 → No), the processing unit 100 advances the processing to step S804.
If failure information is received as a result of step S801 (S801 → Yes), the topology information processing unit 101 updates the topology information based on the received failure information (S802), and updates the topology to the master route server 1401. Information is transmitted (S803). The process for updating the topology information based on the failure information is the same as the process performed in the first embodiment.

Subsequently, the failure / recovery information receiving unit 105 determines whether or not the return information is received from any of the transfer devices 2 in the sub-area 1412 corresponding to the failure / return information receiving unit 105 (S804).
When the return information is not received as a result of step S804 (S804 → No), the processing unit 100 advances the processing to step S807.
If the return information is received as a result of step S804 (S804 → Yes), the topology information processing unit 101 updates the topology information based on the received failure information (S805), and updates the master route server 1401. Topology information is transmitted (S806). The process for updating the topology information based on the return information is the same as the process performed in the first embodiment.

Next, the topology information processing unit 101 determines whether or not the topology information of all areas distributed in step S703 or step S708 in FIG. 17 has been received from the master route server 1401 (S807).
As a result of step S807, when the topology information of all areas has not been received (S807 → No), the processing unit 100 returns the process to step S801.
If the topology information of all areas is received as a result of step S807 (S807 → Yes), the sub-area that the primary route calculation unit 102 is the distribution target of the route information 300 based on the received topology information of all areas The primary route information 301 in 1412 is calculated (S808).
Then, the route setting unit 104 distributes the calculated primary route information 301 to the transfer device 2 in the subarea 1412 that is the distribution target of the route information 300 (S809), and the processing unit 100 returns the process to step S801.

  In the second embodiment, since the processing of the transfer device 2 is the same as that of the first embodiment, illustration and description are omitted.

According to the second embodiment, the route server 1 only has to calculate the route information 300 for the transfer device 2 in the area to which the route server 1 corresponds, and thus the calculation processing load of the route information 300 can be reduced. .
Furthermore, in the first embodiment, it is necessary for one route server 1 to collect all the adjacent information in the network system 10a. However, in the second embodiment, the adjacent information in the area to which the network system 10a corresponds is stored. Since collection is sufficient, communication efficiency can be improved.
Furthermore, since communication between each route server 1 is only topology information, the communication load between each route server 1 can be made light.

In the second embodiment, when a failure occurs in the link of the sub-area 1412, the sub-route server 1402 that has received the failure information calculates the topology information of the sub-area 1412 in which the failure has occurred, and the topology information is obtained. Although transmitted to the master route server 1401, the sub route server 1402 may transmit only the received failure information to the master route server 1401. At this time, the master route server 1401 updates the topology information of all areas based on the received failure information. In this way, the sub route server 1402 only needs to transfer the failure information to the master route server 1401, and the master route server 1401 also updates the connection of the link described in the failure information from “present” to “none”. Therefore, the calculation load of the master route server 1401 and the sub route server 1402 is further reduced than the method described in FIGS. 17 and 18, and the communication load is further reduced than the method described in FIGS. 17 and 18. The
Note that the processing at the time of return can be performed in the same manner.

DESCRIPTION OF SYMBOLS 1 Path | route server 2, 2a-2f, 2o-2q, 2A-2E Transfer apparatus 10, 10a Network system 100 Processing part 101 Topology information processing part 102 Primary path | route calculation part (path | route information calculation part)
103 Backup route calculation unit (route information calculation unit)
104 Route Setting Unit 105 State Change Detection Unit 106 Operator I / F Unit 110 Topology Information DB
120 Primary route information DB
130 Backup route information DB
DESCRIPTION OF SYMBOLS 200 Operation control part 201 Adjacent information transmission part 202 Failure / restoration information transmission part 203 Path | route information reception part 204 Hardware management part 210 Packet control part 211 State change detection part 212 Packet transfer part 220 Primary path | route information DB (memory | storage part)
230 Backup path information DB (storage unit)
300 route information 301 primary route information 302 backup route information 400 packets (communication data)
1401 Master route server 1402, 1402a, 1402b Sub route server 1411 Backbone area (first area)
1412, 1412a, 1412b Sub-area (second area)

Claims (10)

Collect adjacency information, which is information about adjacency relationships between the transfer devices, from the transfer device, calculate route information that is a transfer path of communication data based on the collected adjacency information, and transfer the calculated route information to the transfer device A route server that distributes to each device,
A path information calculation unit for calculating, as the path information, primary path information used before a failure of a path connecting the transfer apparatuses and one or more backup path information used when a failure of the path occurs;
A route setting unit that distributes the calculated primary route information and the backup route information to each of the transfer devices;
A route server characterized by comprising: A transfer device that transfers received communication data with reference to route information stored in its own storage unit,
A storage unit storing primary route information distributed from the route server and the one or more backup route information;
When a failure occurs in the route, a packet control unit that transfers the communication data by switching the route information to be referenced from the primary route information to the backup route information;
A transfer apparatus comprising: The packet control unit
The transfer device according to claim 2, wherein when the communication data is transferred according to the backup path information, if a failure occurs in the path, the received communication data is discarded. A route server collects adjacent information, which is information related to an adjacency relationship between the transfer devices, from a transfer device, calculates route information that is a transfer route of communication data based on the collected adjacent information, and calculates the calculated route Distribute information to each of the transfer devices,
The transfer device is a route control method by a network system that refers to the distributed route information and transfers received communication data,
The route server is
As the route information, primary route information used before failure of the route connecting the transfer devices and one or more backup route information used when the route failure occurs are calculated and distributed to the transfer devices. And
Each of the transfer devices is
The primary route information distributed from the route server and the backup route information are stored in a storage unit,
When a failure occurs in the route, the route data to be referred is switched from the primary route information to the backup route information to transfer the communication data. The network system is divided into a plurality of areas,
It is set which route server distributes the route information to the transfer device belonging to which area,
Each of the plurality of route servers calculates the primary route information and the backup route information to be delivered to the transfer device in the area to which the route information is to be delivered,
5. The route control method according to claim 4, wherein the calculated primary route information and backup route information are distributed to a transfer device in the area that is the distribution target of the route information. The plurality of areas are:
A first area which is a distribution target of route information by a master route server which is one of the plurality of route servers;
A second area which is a distribution target of route information by a sub route server which is a route server different from the master route server;
Have
The sub-route server is
Collecting neighbor information from transfer devices belonging to the second area, calculating topology information, which is information relating to arrangement of transfer devices in the second area, based on the collected neighbor information, and calculating the topology information To the master route server,
The master route server collects adjacency information from transfer devices belonging to the first area, collects topology information of the second area from the sub route server, and adjoins the collected first area. Calculating topology information of all areas including the first area and the second area based on the information and the topology information of the second area;
Deliver the topology information of all areas to the sub-route server,
Each of the master route server and the sub route server calculates route information of the transfer device belonging to the area to which the route information is to be distributed, based on topology information of the entire area. The route control method according to claim 5. The master route server is
When receiving failure information including information on a transfer device in which a failure occurs between transfer devices in the first area,
Based on the topology information of the second area stored in its own storage unit and the received failure information, calculate the topology information of all areas including the first area and the sub-area,
Deliver the topology information of all areas to the sub-route server,
The master route server and the sub route server calculate primary route information of the transfer device belonging to the area to which the route information is to be distributed, based on topology information of the entire area. The routing control method according to 1. The sub-route server is
When receiving failure information including connection information between transfer devices in which a failure occurs between transfer devices in the second area that is the distribution target of the route information,
Send information including the failure information to the master route server,
The master route server is
Based on the topology information of the second area stored in its own storage unit and the transmitted failure information, the topology information of all areas including the first area and the sub-area is calculated. ,
Deliver the topology information of all areas to the sub-route server,
The master route server and the sub route server calculate primary route information of the transfer device belonging to the area to which the route information is distributed based on topology information of the entire area. Item 7. A path control method according to Item 6. A route server collects adjacent information, which is information related to an adjacency relationship between the transfer devices, from a transfer device, calculates route information that is a transfer route of communication data based on the collected adjacent information, and calculates the calculated route Distribute information to each of the transfer devices,
The transfer device is a network system for transferring received communication data with reference to the distributed route information,
The route server is
As the route information, primary route information used before failure of the route connecting the transfer devices and one or more backup route information used when the route failure occurs are calculated and distributed to the transfer devices. And
Each of the transfer devices is
The primary route information distributed from the route server and the backup route information are stored in a storage unit,
When a failure occurs in the path, the communication data is transferred by switching the path information to be referenced from the primary path information to the backup path information. The network system is divided into a plurality of areas,
It is set which route server distributes the route information to the transfer device belonging to which area,
Each of the route servers calculates the primary route information and the backup route information for the transfer device in the area,
The network system according to claim 9, wherein the calculated primary route information and backup route information are distributed to a transfer device in the area to which the route information is to be distributed.

Images