JP2005223522A - Path calculation method, path calculation control apparatus, and path calculation program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a path calculation method for designing a network wherein a superordinate layer can reliably be restored even on the occurrence of a fault in a subordinate layer link while enhancing a traffic accommodation efficiency in the network wherein one or more user groups exist. <P>SOLUTION: The path calculation method is used for the network wherein at least two routers having functions of traffic transfer or the like and connected to switches of the subordinate layers and a plurality of user groups exist in the network having the superordinate and subordinate layers. Paths are set wherein prescribed information of the superordinate layer for only users belonging to an objective user group is taken into account, the traffic accommodation efficiency is excellent only between users belonging to the same user group on the basis of the information, any layer existing in the network copes with a fault by any method on the occurrence of the fault in the subordinate layer and the superordinate layer can reliably be restored. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a route calculation method, a route calculation control device, and a route calculation program, in particular, a packet network such as MPLS (Multi Protocol Label Switching) and ATM (Asynchronous Transfer Mode), The present invention relates to a route calculation method considering both TDM (Time Division Multiplexing) and WDM (Wavelength Division Multiplexing) optical networks.

In recent years, IP traffic has increased explosively. In the conventional connectionless system that carries traffic via a large number of routers, there is a problem that the equipment cost of the core network that transfers increasing IP traffic increases.
Therefore, there is a method for increasing the capacity of traffic by passing through IP (Layer 3) and Ethernet (Layer 2) traffic in a connection-type network such as wavelength and TDM, and as a result, realizing the economy of the network. Appeared.
A core network that realizes a large-capacity IP traffic network service economy by such a method has also been proposed (for example, see Non-Patent Document 1).
In a connection-type network that establishes a connection and passes through a connectionless network that carries traffic via a large number of routers, it is necessary to establish a path before the traffic flows. Must be calculated in advance.
In order to set a path that can efficiently handle a large amount of information by efficiently using a limited bandwidth, a server or router in the network is provided with a Constrained Shortest Path First (CSPF) function.

The CSPF function is a function that calculates a path of a path that satisfies a constraint based on a request from the outside such as a module that exchanges signaling information and answers the path to the outside, that is, a constrained shortest path calculation function.
To date, a great many CSPF algorithms have been proposed. Since each CSPF algorithm has different functions and performance, it is necessary to select an appropriate one according to the required constraints.
If the CSPF algorithm to be used is different, various factors such as network cost, traffic accommodation efficiency, and reliability for end users are affected. In other words, by using an appropriate CSPF algorithm, it is possible to design a path that meets the purpose, such as a reduction in network construction cost.
The CSPF algorithm differs depending on the attribute / constraint of the set path, and the optimum algorithm varies depending on the layer such as an IP network or an optical network.
In an optical network, a wavelength assignment algorithm is required in addition to a route calculation algorithm.
Typical path calculation algorithms include k-shortest path heuristic and linear programming (see, for example, Non-Patent Document 2).

The conventional CSPF algorithm is closed to one layer such as only an IP network and only an optical network, and performs route calculation considering only one layer.
However, in a core network that achieves economics by performing pass-through as described above, path switching with good bandwidth utilization efficiency is possible if different CSPF algorithms are applied to the upper layer and lower layer and route calculation is performed independently. Cannot be performed, and an efficient network cannot be constructed.
Furthermore, an optical network has a link in which a plurality of optical fibers are bundled, and if only such a link is set as a route, a large amount of communication is interrupted when this link fails. End up.
Therefore, in order for the path of the upper layer to pass through a path that does not concentrate on a specific link in the lower (physical) layer and does not overlap each other, it is necessary to calculate a path while considering the two layers with each other.
Under such circumstances, the necessity of considering both the upper layer and the lower layer has begun to be recognized, and recently, a multi-layer CSPF algorithm capable of performing route calculation in consideration of both the upper layer and the lower layer Began to attract attention.
Here, the multi-layer refers to a network in which a connection type network such as a packet network is accommodated in a connection type network such as a circuit switching network.

Two typical CSPFs that take into account the multilayer are listed below.
(1) BXCQ (Branch Change with Quality-of-Service Constraint) method It is intended to perform dynamic route control in consideration of the traffic volume of the upper layer path and the lower layer.
By exchanging upper layer path traffic information between nodes, the topology of the lower layer path is set based on that information and dynamically changed, enabling the construction of a network that can adapt to changes in traffic demand. Become. The BXCQ method is used for actual route calculation.
This CSPF algorithm is effective when applied to a case where it is desired to set a path with priority given to the traffic accommodation efficiency of the network (see, for example, Non-Patent Document 3).
(2) SLA (Survivable Layout Algorithm) method Even if a failure occurs in an arbitrary link in the lower layer, each node in the upper layer prepares a certain detour route that is not affected by the failed link in advance. The purpose is to design a path that maintains the connection and ensures the connection of the upper layer network.
For each link in the lower layer, investigate the effect of the upper layer when a failure occurs, and recalculate the route if necessary. A network that can be restored can be constructed. The SLA method is used for actual route calculation.
This algorithm is effective when applied to a case where it is desired to set a path with a high priority of reliability that guarantees connection between nodes even if a failure occurs in the network (for example, Non-patent document 4).

Takashi Kurimoto, Takashi Miyamura, Michihiro Aoki, Ichiro Inoue, Nobuaki Matsuura, Hisashi Kojima, Shigeo Urushitani, “Proposal of Multi-layer Service Network Architecture”, Technical Report of IEICE, PN2003-6 (2003-09). R. Bhandari, Survival Networks: Algorithms for Diverse Routing, Kluwer Academic Publishers, 1999. Kohei Shiomoto, Eiji Oki, Masaru Katayama, Wataru Imajuku and Naoaki Yamanaka, “Dynamic multi-layer traffic engineering in GMPLS networks” ISS2002. Galen H. Sasaki and Ching-Fong SuInterface, `` Between IP and WDM and its Effect on the Cost of Survivability, '' IEEE Communication Magazine, vol.41, no.1, pp.74-79, January, 2003.

However, the above-mentioned CSPF that takes into account the multi-layer has a problem when applied to an actual network.
For example, in the case of the BXCQ method, the path is established in consideration of the traffic amount of the upper layer, so that the traffic accommodation efficiency increases and an economical network can be realized.
However, since there is no consideration for recovery methods in the event of a link failure in the lower layer, many communications will be interrupted or a dedicated protection path will be prepared if a failure actually occurs in the network. There is a problem that it is costly because it is necessary to do so.
In the SLA method, which is another multi-layer CSPF, even if a failure occurs in any link in the lower layer, nodes in the upper layer prepare in advance some sort of detour route that is not affected by the failed link. Thus, the reliability of the upper layer network can be provided by maintaining the connection.
However, because the amount of traffic between nodes is not taken into consideration, there is a problem that it is not possible to establish a path with good traffic accommodation efficiency, a delay occurs when carrying traffic, and the cost for establishing a path is high. Come out.

A problem common to both algorithms is that a case where a plurality of user groups exist in the network is not assumed.
If there is an environment in which the capacity of the network increases and services can be provided to many users, there may be a situation where there are multiple groups such as between the same company and within the same region.
In this case, users belonging to the same user group communicate with each other, but do not communicate with users belonging to different user groups. Furthermore, it must be ensured that user information in the group does not leak to another user group.
Therefore, there is a need for a method for grasping user information in the network and establishing a path only between users belonging to the same user group.
The conventional multi-layer CSPF mentioned above is not compatible with such a network in which a plurality of users exist.

The present invention has been made to solve the above-described problems of the prior art, and an object of the present invention is to improve traffic accommodation efficiency in a network where one or more user groups exist, while lower layer. An object of the present invention is to provide a route calculation method and route calculation control device for designing a network so that an upper layer can be reliably recovered by some method even when a failure occurs in a link.
Another object of the present invention is to provide a route calculation program for causing a computer to execute the route calculation method described above.
The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

Of the inventions disclosed in this application, the outline of typical ones will be briefly described as follows.
In order to solve the above-described problems, in the present invention, there are at least two or more routers having a function of traffic forwarding in a network in which an upper layer and a lower layer exist, and connected to a lower layer switch. In the route calculation method when there are multiple user groups, by considering the predetermined information of the upper layer only for users belonging to the target user group, only between users belonging to the same user group based on that information When a failure occurs in the lower layer with good traffic accommodation efficiency, the failure is handled in some way in any layer in the network, and the path is set so that the upper layer can reliably recover It is characterized by that.
In the present invention, the lower layer includes a wavelength or TDM connection-type network, and the upper layer includes an MPLS connection-type network.
In the present invention, the lower layer includes a wavelength or TDM connection type network, and the upper layer includes a packet network connectionless network.

Further, in the present invention, when a failure occurs in a lower layer, both the upper layer and the lower layer can cope with the failure by a recovery method possible in each layer, and the upper layer can be reliably recovered. Such a path setting is performed.
In the present invention, when a failure occurs in the lower layer, the failure is dealt with by a recovery method that is possible only in the upper layer, and path setting is performed so that the upper layer can be reliably recovered. And
The present invention is characterized in that path setting is performed using a traffic matrix as upper layer information.
Further, the present invention is characterized in that a path is set using future traffic demand prediction as upper layer information.

Also, in the present invention, the link cost is used as the information of the upper layer, and the path setting is performed in consideration that the upper layer path does not use the same lower layer link as a condition for the link cost. It is characterized by.
Further, in the present invention, traffic accommodation efficiency is good by considering predetermined information of the upper layer, and when a failure occurs in the lower layer, the failure is dealt with in some way in any layer existing in the network. After setting the path so that the upper layer can be reliably restored, set the path for the upper layer again based on the lower layer path set at this time, and minimize the delay. It is characterized by performing.
In the present invention, the path calculation control device that executes the above-described path setting method is used exclusively for the purpose of centrally managing and controlling each node in the network.
In the present invention, the route calculation control apparatus that executes the path setting method described above manages and controls the nodes attached to each node in the network.
The present invention is also a route calculation program for causing a computer to execute the route calculation method described above.

The effects obtained by the representative ones of the inventions disclosed in the present application will be briefly described as follows.
According to the present invention, an upper layer path is accommodated in a lower layer path, and in a network in which the upper layer and the lower layer cooperate with each other, bandwidth utilization efficiency is good. A highly reliable route design in which failure recovery can be performed by performing routing becomes possible.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
In all the drawings for explaining the embodiments, parts having the same functions are given the same reference numerals, and repeated explanation thereof is omitted.
[Example 1]
First, a network structure to which the route calculation control apparatus according to the first embodiment of the present invention is applied will be described.
FIG. 1 is a conceptual diagram of a network to which the route calculation control device of this embodiment is applied, and is a conceptual diagram of a network in which a packet network is accommodated in a circuit switching network.
The network shown in FIG. 1 includes switches (A1, B1, C1, D1), routers (A2, B2, C2, D2), and a route calculation control device 10.
Each switch (A1, B1, C1, D1) has a role of transferring wavelength and TDM data. Each switch is connected by a physical link to form a circuit switching network.
Each router (A2, B2, C2, D2) is connected to the switch (A1, B1, C1, D1) and forwards the packet by passing data to each switch (A1, B1, C1, D1) It has a function to do. Each router is connected by a logical path provided by setting a lower layer path, and constitutes a packet network.
When there are a plurality of user groups in the network, the route calculation control apparatus 10 collects some information only for users belonging to the target user group, and only between users belonging to the same user group based on the information. It has a function to calculate the path for path design, and exchanges information necessary for path setting with each router.

In the network as shown in FIG. 1, the upper layer is accommodated in the lower layer, and the lower layer provides a virtual topology to the upper layer.
The upper layer means a connectionless type or connection type network accommodated in a network relatively lower than the own network, and the lower layer means a network relatively higher than the own network. It means a connection type network to accommodate.
The upper layer is an IP or Ethernet network or an MPLS network that transfers IP packets with labels, and the lower layer is a TDM or WDM network.
Next, information required in this embodiment will be defined.
(1) Link cost A cost set based on the distance or the number of hops necessary to carry traffic. Usually, the link cost increases as the distance and the number of hops increase.
(2) Remaining bandwidth The amount of bandwidth remaining in the lower layer link. For optical paths, the number of wavelengths available.
(3) Traffic matrix Information on the amount of traffic exchanged between higher layer nodes.

FIG. 2 is a block diagram showing a schematic configuration of the route calculation control apparatus 10 shown in FIG.
As shown in FIG. 2, the route calculation control device 10 includes a route calculation unit 11, a path information storage unit 12, a routing information storage unit 13, a membership information storage unit 14, an upper layer information storage unit 15, A path information control unit 16, a control interface 17 with an external device, and a routing information control unit 18 are provided.
The table of FIG. 3 shows how the functions of each unit and the information necessary to implement the above-described algorithm correspond to the function blocks.
(1) The route calculation unit 11 performs network path calculation.
(2) The path information storage unit 12 collects and stores upper / lower layer path information in the network, lower layer link information used by the upper layer path, and upper layer path traffic information.
(3) The routing information storage unit 13 stores upper / lower layer topology information and link cost / residual bandwidth information.
(4) The membership information storage unit 14 manages and stores user information.
(5) The upper layer information storage unit 15 creates upper layer information for each user from the information in the path information storage unit 12 and the membership information storage unit 14 and stores it in the internal database 151.
(6) The path information control unit 16 controls the flow of path establishment in the upper layer and the lower layer.
(7) The control interface 17 with the external device sends a path establishment / cancellation command and accompanying information to the external device (in this case, a router).
(8) The routing information control unit 18 collects information for attaching a cost to a link in the network, and transmits link cost information necessary for route calculation to the routing information storage unit 13.

In the conventional multi-layer route calculation algorithm, the lower layer path is established in consideration of the upper layer traffic, and the upper layer network is kept connected even when any link of the lower layer is disconnected. There was something to do the path design.
However, these algorithms are specialized to either consider traffic accommodation efficiency or guarantee upper layer connectivity in the event of a failure, and there is a problem that there is no algorithm that satisfies both .
In the present embodiment, when performing route calculation in consideration of multi-layers, the route calculation is performed based on predetermined information of the upper layer to increase the traffic accommodation efficiency and at the same time, even when a failure occurs in the lower layer By performing path design in advance so as to ensure layer connectivity, the above-mentioned problems are solved.
FIG. 4 is a flowchart for explaining the route calculation method according to the embodiment of the present invention, and is a flowchart for explaining the route determination operation of the multilayer network.
First, based on predetermined information of the upper layer, it is determined which node the lower layer path is to be established between, and the lower layer topology is determined (step 101).
Here, as the information of the upper layer, a traffic matrix, a future traffic demand prediction, a usage fee paid by a network user, a link cost weighted under some condition, or the like is used.
Next, the upper layer topology is determined by accommodating the upper layer path in the lower layer path determined in step 101 (step 102).

At this time, the route may be determined using the restricted Dijkstra algorithm, or the link through may be explicitly specified. Moreover, you may devise how to attach link cost well so that the link used by each high-order path may not overlap.
By devising the way of attaching links in this way, the upper layer path is distributed without concentrating on specific links in the lower layer, so even if a failure occurs in the lower layer link, the damage range due to communication interruption Is suppressed.
Further, the number of paths to be added can be suppressed at the stage of checking the connectivity of the next step. Details on how to set this cost will be described later.
Further, even if a failure occurs in any lower layer link, it is checked with respect to the lower layer link whether the upper layer connection is maintained (step 103).
At this time, all lower layer physical links may be checked, or only specific physical links may be checked.
In this way, considering the case where a failure occurs for a specific physical link for which all or important ones have been selected, the possible lower layer failures are considered to the maximum, It is possible to improve the upper layer reliability when a failure occurs.
Finally, if necessary, re-calculate the lower layer route for the upper layer path, or add a lower layer path, etc. to secure a detour route that always maintains the upper layer connection. (Step 104).
As a result of such a route calculation, a topology is finally completed so that traffic accommodation efficiency is good and upper layer connectivity is ensured even when a lower layer failure occurs.
Note that after step 104, the upper layer path design may be performed again based on the lower layer path set in each of the above steps, and the path setting may be performed so as to minimize the delay.

Details of this route calculation algorithm will be described next.
In order to determine the topology of the upper layer and the lower layer, predetermined information of the upper layer is used.
FIG. 5 is a flowchart for explaining an algorithm for determining path paths of upper and lower layers after collecting predetermined information of upper layers in the route calculation method of the embodiment of the present invention.
The upper layer information used at this time provides a service from a plurality of user groups existing in the network using the user group information stored in the membership information storage unit 14 of the route calculation control device 10. Collect and use only the information of users belonging to the target group.
As the upper layer information, a traffic matrix, a future traffic demand prediction, a usage fee paid by a network user, or a link cost weighted under some condition is used.
First, based on predetermined information of the upper layer, it is determined which node the lower layer path is to be established between, and the lower layer topology is determined (step 111).
At this time, by devising the cost attached to the lower layer link, it is possible to avoid the concentration of the lower layer path on a specific physical link. As a result, the possibility that the upper layer path passes through the same lower layer link is reduced. The upper layer reliability at the time of a layer failure can be improved.

There are two examples of how to add this cost.
(1) Cost example 1
First, when a single lower layer path is established, a cost is added to the lower layer link directly connected to the start / end node of the path. When the second lower layer path is finished, the same cost is added again.
By repeating this, there is a high possibility that the path of the next path to be stretched will not overlap with the path stretched so far. Therefore, it is possible to avoid the concentration of lower layer paths on a specific physical link.
(2) Cost example 2
First, when a path of one lower layer is established, a cost is attached to all the lower layer links through which this path passes.
The same cost is applied to the lower layer links for the second and subsequent lower layer paths. By repeating this, there is a high possibility that the path of the next path to be stretched will not overlap with the path stretched so far.
Therefore, it is possible to avoid the concentration of lower layer paths on a specific physical link.

Next, in order to accommodate the upper layer path in the lower layer path, the shortest path of the lower layer path through which each upper layer path passes is calculated (step 112).
This calculation is performed by the route calculation unit 11 of the route calculation control device 10.
At this time, the lower layer route is calculated in an order from an upper layer path with an earlier order by determining some order in the upper layer path. The order is determined based on the traffic volume and the usage fee paid by the user. Dijkstra Algorithm can be used to calculate the shortest path.
When attaching a cost, you may consider the information regarding the residual bandwidth and distance stored in the routing information storage unit 13 of the route calculation control device 10 and the fact that the upper layer path does not pass through the same lower layer path as much as possible.
At this time, by devising the cost attached to the lower layer path, it is possible to avoid the concentration of the upper layer path in a specific lower layer path, and as a result, the possibility that the upper layer path passes through the same lower layer path is reduced. The upper layer reliability at the time can be improved.
As a method of assigning a cost so that the upper layer path does not pass through the same lower layer path as much as possible, a cost is added to the lower layer path that accommodates the upper layer path after first establishing one upper layer path. The same cost is applied to the lower layer path for the second and subsequent upper layer paths.
By repeating this, there is a high possibility that the path of the next path to be stretched will not overlap with the path stretched so far.
Therefore, it is possible to avoid the upper layer path from being concentrated on a specific lower layer path.

Next, it is determined whether or not there is an available bandwidth that can flow traffic on the shortest route that is found (step 113).
Available bands include wavelengths and time slots. When assigning wavelengths, wavelength assignment algorithms such as the fixed-order method and the Max-Sum method are used.
In step 113, if there is no available bandwidth, it is determined whether there is another lower layer path route through which the upper layer path passes (step 115). In step 115, another route of the lower layer path through which the upper layer path passes. If there is, the shortest path of the next optimal lower layer path is calculated (step 117), and the processing from step 113 is repeated.
In step 115, when there is no lower layer path or other route through which the upper layer path passes, there is no path that directly connects the nodes, so that the target path is abandoned (step 116), and the next upper layer Is selected (step 118), and the processing from step 112 is repeated for the next upper layer path.
On the other hand, if the bandwidth exists in step 113, it is determined whether or not the shortest path of the lower layer has been calculated for all paths of the upper layer (step 114), and the calculation is completed for all upper layer paths. If not, the next upper layer is selected (step 118), and the processing from step 112 is repeated for the next upper layer path.
In step (114), when the shortest path of the lower layer path is calculated for all upper layer paths, the process ends.
The above is the details of the method for determining the connection path of the lower layer that provides the upper layer path. At this time, the same calculation may be performed using MILP (Mixed Integer Linear Program) method.

Next, a description will be given of a path design algorithm that ensures the connection of an upper layer router even when an arbitrary link in the lower layer is disconnected.
FIG. 6 is a flowchart for explaining this algorithm.
First, after the upper and lower layer topologies are completed by the method described with reference to FIG. 5, the lower layer links are ordered (step 121).
This order may be manually assigned in advance before the path is established.
In accordance with this order, lower layer paths using the lower layer links are examined, and a lower layer topology in which those paths are deleted is created (step 122).
At this time, it is determined whether or not there is a node isolated from the network (step 123). If there is no node isolated from the network in step 123, the next lower layer link is selected (step 129), and The processing from step 122 is repeated for the next lower layer link.
Here, the node isolated from the network means a node that does not have any upper layer path for exchanging traffic with other nodes in the same upper layer network.

If there is an isolated node from the network, create a lower layer topology that excludes the lower layer link, and whether there is a lower layer path that connects the isolated node to other nodes (Step 124), and if there is no node that can connect the isolated node to the network in Step 124 (that is, if an appropriate path cannot be found), the connectivity of the network to the physical link is guaranteed. If not, the next lower layer link is selected (step 129), and the processing from step 122 is repeated for the next lower layer link.
Such a path may be dealt with by preparing a new 1 + 1 protection path after it is found that the connectivity of the network cannot be maintained as a result of the calculation.
If there is a lower layer path connecting the isolated node and other nodes in step 124, the lower layer path may be the lowest cost or the one found first.
If there is a lower layer path connecting the isolated node and other nodes in step 124, it is determined whether there is a usable bandwidth in the lower layer path (physical link) (step 125). .
When checking whether there is a wavelength resource, a wavelength allocation algorithm such as the fixed-order method or the Max-Sum method is used.

If there is no available bandwidth in the lower layer route at step 125, it is checked whether there is another lower layer route (step 130). If there is another lower layer route at step 130, Returning to step 125, similarly, it is checked whether there is an available bandwidth.
If there is no bandwidth of the lower layer path at step 125 and there is no other lower layer path at step 130, the next lower layer link is selected without guaranteeing the network connectivity to the physical link. (Step 129), the process from Step 122 is repeated for the next lower layer link.
Such a path may be dealt with by preparing a new 1 + 1 protection path after the calculation shows that the network connectivity is not maintained.
If there is a usable bandwidth in step 125, the lower layer route of the upper layer path is changed or a new route is prepared as a 1 + 1 protection path while leaving the original route. (Step 126).
At this time, the lower layer link that causes the route change may be stored, and calculation may be performed so that the lower layer link is not used when the same path is subjected to the route change thereafter.
A lower layer path topology obtained by adding the path added in step 126 to the lower layer path topology created in step 122 is created. Based on this topology, it is determined whether there is a node isolated from the network (step 127). If there is still an isolated node, it is checked whether another lower layer route exists (step 130).
If there is no node isolated from the network at step 127, it is determined whether the above calculation has been performed for all links in the lower layer or a specific link to be examined (step 128). If there is a lower layer link that is not considered, the processing from step 122 is repeated for the next lower layer link.
If all physical layer links are considered in step 128, the process ends.
The above is the details of the algorithm.

The route calculation method described above can also be executed by a computer. In this case, the route calculation method stored in a hard disk or the like in the computer is executed by the computer.
The route calculation program is supplied by a recording medium such as a CD-ROM or a transmission medium having a function of transmitting information such as a network such as the Internet or a communication line such as a telephone line.
The route calculation program may be for realizing a part of the above-described processing. Furthermore, what can implement | achieve the process mentioned above in combination with the program already recorded on another apparatus, what is called a difference file (difference program) may be sufficient.
As described above, according to the present embodiment, the upper layer path is accommodated in the lower layer path, and the upper layer path is established based on the predetermined information of the upper layer in the network that cooperates with the upper layer and the lower layer. However, if there is a failure in any lower layer link, the route is recalculated if necessary so that the connectivity of the upper layer is maintained. Therefore, it is possible to design a route with high reliability that failure recovery can be performed by performing rerouting in an upper layer when an error occurs.
In this embodiment, it is assumed that both the upper layer and the lower layer have a failure recovery function. By this method, it is possible to realize a network that is economical and allows the end user to continue using the network without worrying about obstacles.

[Example 2]
In the first embodiment, the above-described route calculation method and route calculation program are applied to a network in which a plurality of users are accommodated.
This route calculation method can also be applied when all nodes in the network belong to the same group. In this case, the route calculation control device 10 is configured as shown in FIG.
As illustrated in FIG. 7, the route calculation control device 10 according to the present exemplary embodiment includes a route calculation unit 11, a path information storage unit 12, a routing information storage unit 13, an upper layer information storage unit 15, and a path information control unit. 16, a control interface 17 with an external device, and a routing information control unit 18.
The table of FIG. 8 shows how the functions of each unit and the information necessary to implement the above-described algorithm correspond to the function blocks.

(1) The route calculation unit 11 performs network path calculation.
(2) The path information storage unit 12 collects and stores upper / lower layer path information in the network, lower layer link information used by the upper layer path, and upper layer path traffic information.
(3) The routing information storage unit 13 stores upper / lower layer topology information and link cost / residual bandwidth information.
(4) The upper layer information storage unit 15 creates upper layer information for each user from the information in the path information storage unit and the membership information storage unit, and stores it in the internal database 151.
(5) The path information control unit 16 controls the flow of path establishment in the upper layer and the lower layer.
(6) The control interface 17 with the external device sends a path establishment / cancellation command and accompanying information to the external device (in this case, a router).
(7) The routing information control unit 18 collects information for attaching a cost to a link in the network, and transmits link cost information necessary for route calculation to the routing information storage unit 13.
By using only one user group as in the present embodiment, the membership information storage unit 14 of the route calculation control device 10 of the first embodiment is not necessary, and a plurality of higher layers that are necessary Since only one database 151 in the information storage unit 15 is required, the implementation can be simplified.

[Example 3]
In the first embodiment and the second embodiment described above, the route calculation control device 10 is used exclusively for the purpose of centrally managing and controlling each node in the network collectively.
In the present embodiment, the route calculation control device 10 is an embodiment arranged inside each router.
FIG. 9 is a conceptual diagram of the network of this embodiment. As shown in FIG. 9, in this embodiment, the routers (A2, B2, C2, D2) are routers with a route calculation function.
As described above, in this embodiment, the route calculation control device is arranged inside each router (A2, B2, C2, D2), and manages and controls the node attached to each node in the network.
In the present embodiment, the route calculation method and the route calculation control device 10 can use the same ones as those in the first embodiment and the second embodiment. Similarly, the route calculation method in the present embodiment uses the route calculation method. It is also possible for a computer to execute the program.
As mentioned above, the invention made by the present inventor has been specifically described based on the above embodiments. However, the present invention is not limited to the above embodiments, and various modifications can be made without departing from the scope of the invention. Of course.

1 is a conceptual diagram of a network to which a route calculation control device according to a first embodiment of the present invention is applied. It is a block diagram which shows schematic structure of the route calculation control apparatus of Example 1 of this invention. It is a figure explaining the information which each part in the path | route calculation control apparatus of Example 1 of this invention handles, and the function of each part. It is a flowchart for demonstrating the route calculation method of Example 1 of this invention. 6 is a flowchart for explaining an algorithm for determining path routes of upper and lower layers after collecting predetermined information of upper layers in the route calculation method according to the first embodiment of the present invention. FIG. 6 is a flowchart for explaining a path design algorithm that ensures connection of a router in an upper layer even when an arbitrary link in a lower layer is disconnected in the route calculation method according to the first exemplary embodiment of the present invention. It is a block diagram which shows schematic structure of the path | route calculation control apparatus of Example 2 of this invention. It is a figure explaining the information which each part in the path | route calculation control apparatus of Example 2 of this invention handles, and the function of each part. It is a conceptual diagram of the network to which the path | route calculation control apparatus of Example 3 of this invention is applied.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Path | route calculation control apparatus 11 Path | route calculation part 12 Path information storage part 13 Routing information storage part 14 Membership information storage part 15 Upper layer information storage part 16 Path information control part 17 Control interface with an external apparatus 18 Routing information control part 18
151 Database A1, B1, C1, D1 Switch A2, B2, C2, D2 Router

Claims (20)

When a connectionless type or connection type network accommodated in a network relatively lower than the own network is an upper layer, and a connection type network accommodating a network relatively higher than the own network is a lower layer, There is an upper layer and a lower layer, and there are at least two upper layer routers connected to the lower layer switch. The network exchanges traffic only with users belonging to the same group in the network, A route calculation method that designs paths only between users belonging to the same user group in a network where there is at least one user group defined not to exchange traffic with users belonging to the group.
Determining which lower layer path to establish between which nodes based on predetermined information of the upper layer, and determining a lower layer topology;
Step 2 of accommodating the upper layer path in the lower layer path determined in Step 1 and determining the topology of the upper layer;
A route calculation method comprising: step 3 of checking a lower layer link whether or not an upper layer connection is maintained even when a failure occurs in an arbitrary lower layer link.   2. The route calculation method according to claim 1, wherein when only one user group exists in the network, predetermined path information is collected for all users in the user group and path setting is performed. . The lower layer includes a wavelength or TDM connection type network,
The route calculation method according to claim 1, wherein the upper layer includes an MPLS connection-type network. The lower layer includes a wavelength or TDM connection type network,
The route calculation method according to claim 1, wherein the upper layer includes a connectionless network of a packet network.   If a failure occurs in a lower layer, both the upper and lower layers handle the failure using the recovery methods available in each layer, and set the path so that the upper layer can reliably recover. The route calculation method according to any one of claims 1 to 4, wherein the route calculation method is provided.   2. If a failure occurs in a lower layer, the failure is dealt with by a recovery method that is possible only in the upper layer, and path setting is performed so that the upper layer can be reliably recovered. The route calculation method according to claim 4.   The route calculation method according to claim 1, wherein the upper layer information is a traffic matrix.   The route calculation method according to claim 1, wherein the information of the higher layer is a future traffic demand prediction. The upper layer information is a link cost,
The path design is performed in consideration of not using the same lower layer link as a higher layer path as a condition that becomes a reference of the link cost. Route calculation method.   In step 3, when a lower layer link is checked whether or not an upper layer connection is maintained even if a failure occurs in any lower layer link, the processing is performed for all lower layer physical links. The route calculation method according to any one of claims 1 to 9, wherein the route calculation method is characterized.   In the step 3, when a lower layer link is checked whether or not an upper layer connection is maintained even when a failure occurs in any lower layer link, it is performed only for a specific physical link. The route calculation method according to any one of claims 1 to 9.   After the step 3, there is a step 4 in which the path design of the upper layer is performed again based on the lower layer path set in the steps 1 to 3, and the path setting for minimizing the delay is performed. The route calculation method according to any one of claims 1 to 11.   A route calculation program for causing a computer to execute the route calculation method according to any one of claims 1 to 12. When a connectionless type or connection type network accommodated in a network relatively lower than the own network is an upper layer, and a connection type network accommodating a network relatively higher than the own network is a lower layer, There is an upper layer and a lower layer, and there are at least two upper layer routers connected to the lower layer switch. The network exchanges traffic only with users belonging to the same group in the network, A route calculation control device that performs path design only between users belonging to the same user group in a network where there is at least one user group defined not to exchange traffic with users belonging to the group. Te,
A route calculator that performs network path calculations;
An upper layer information storage unit that stores upper layer information for each user;
The route calculation unit determines, based on predetermined information of an upper layer stored in the upper layer information storage unit, which node establishes a path of a lower layer, and determines a lower layer topology; ,
Means 2 for accommodating the upper layer path in the lower layer path determined by the means 1, and determining the topology of the upper layer;
A route calculation control device comprising: means 3 for checking whether or not an upper layer connection is maintained even when a failure occurs in an arbitrary lower layer link.   15. The route calculation control device according to claim 14, wherein the upper layer information is a traffic matrix.   15. The route calculation control device according to claim 14, wherein the information of the higher layer is a future traffic demand prediction. The upper layer information is a link cost,
15. The path calculation control apparatus according to claim 14, wherein path design is performed in consideration that the upper layer path does not use the same lower layer link as a condition for the link cost.   The route calculation unit includes means 4 for performing path design of the upper layer again based on the lower layer path set by the means 1 to means 3 and setting the path so that the delay is minimized. The route calculation control device according to any one of claims 14 to 17.   The route calculation control device is connected to each router in the network, and centrally manages and controls each node in the network, according to any one of claims 14 to 18. The route calculation control device described.   The route calculation control device is arranged in each router in the network, and manages and controls the node attached to each node in the network. The route calculation control device described.

Images