JP4255080B2 - Network failure recovery management method and network failure recovery management device - Google Patents

Description

  The present invention relates to a multiple customer accommodation network failure recovery management method and a multiple customer accommodation network failure recovery management apparatus. In particular, when a path is set between bases and a plurality of customers are accommodated in the same network, a multi-customer accommodating network failure recovery management method for setting a working and backup path for each customer according to a resource allocation policy, and The present invention relates to a multi-customer accommodation network failure recovery management device.

  As an example of a network (communication network) that accommodates a plurality of customers in the past, an MPLS (Multi Protocol Label Switching) LSP (Label Switched Path) is set in the network, and an L3 VPN (Layer 3 Virtual) that accommodates a plurality of customers is set. Private Network (Private Network) and L2VPN (Layer 2 Virtual Private Network). Further, as an example of accommodating a plurality of customers by setting a wavelength path or a TDM (Time Division Multiplexing) path between customer apparatuses, there is an L1VPN (Layer1 Virtual Private Network).

  It should be noted that “customer” means an administrator for a network operated by a certain common policy. Further, the customer and the manager of the provider who provides the communication path to the customer may be different as usual or may be the same. In the following description, description will be made on the assumption that one customer manages one Virtual Private Network (hereinafter referred to as VPN) in the network.

  As a conventional network management technique, there is one that adds an identifier for each link (Non-Patent Document 1). By using the identifier provided by this technology, it is possible to indicate which customer can use each link, whether the link is in use or backup.

  In addition, as a conventional network management technique, an identifier indicating which failure can occur for a link is added in consideration of the possibility of multiple link failures occurring for one failure. (Non-Patent Document 2). With this technology, route calculation and path setting are performed for multiple working paths that do not fail at the same time for any single failure, taking into account that the resources of the protection path are shared. be able to.

Patent Document 1 describes a technology that provides effective use of network resources and a differentiation service to customers by making proposals to customers after paying attention to customers and considering network topology and network resource status. ing.
Japanese Patent Laying-Open No. 2004-260567 (paragraphs 0015 to 0017, FIG. 2) “Traffic Engineering (TE) Extensions to OSPF Version 2”, IETF, [online], published September 2003, [searched July 2004], Internet <http://www.ietf.org/rfc/rfc3630.txt> “Routing Extension of Support of Generalized Multi Protocol Label Switching”, IETF, [online], October 2003, [July 2004 search], Internet <http://www.ietf.org/internet-drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts/drafts ietf-ccamp-gmpls-routing-09.txt>

  However, in the conventional technology, when there are a plurality of customers, it is not possible to sufficiently define how to behave at the time of failure recovery. Specifically, since it is not possible to specify whether or not backup paths of different customers can share resources, a single failure can be handled when sharing backup path resources with different customers. However, if a plurality of failures occur and a backup path of another customer is set first, there arises a problem that the backup path cannot be used due to insufficient resources.

  On the other hand, if you do not share backup path resources with different customers, even if multiple failures occur, the backup paths of other customers will be set first, making the backup path unusable. However, a backup path can always be secured, but more resources are required. How resources should be used at the time of failure recovery is determined by the circumstances of the customer, and it is not sufficient to determine the response uniformly.

  Therefore, an object of the present invention is to make it possible to perform resource allocation in the event of a failure in a detailed manner corresponding to the customer's request.

In order to solve the above-mentioned problem, in the present invention (Claim 1), in a multiple customer accommodation network in which a plurality of customers are accommodated in the same network, the network failure recovery management device allocates resources of each link to which customer. A resource allocation unit having a function to set whether to allocate the link to a working path or a protection path, a policy setting unit to set a resource allocation policy for each customer, and a resource available to each customer A working path route calculating unit that calculates a path of a working path using a resource, a backup path route calculating unit that calculates a route of a backup path using resources available to each customer, and the working path route calculating unit or switching of the the backup path route calculating unit path setting means for setting a computed path by path when network failure occurs Path switching means, and the resource assignment means assigns the resource to a working path or backup path after assigning the resource to a specific customer or shared resource, respectively, and the policy setting means includes a resource for each customer. The working path route calculating means calculates the working path, the protection path route calculating means calculates the protection path, and the path setting means connects the working path and the protection path to the network. When the network failure of the working path occurs, the path switching means switches the path to the preset backup path.

  According to this configuration, it is possible to perform resource management and network failure recovery that precisely respond to customer requests.

  In the present invention (Claim 2), the resource allocation means allocates link resources to one or more specific customers or arbitrary customers, and which customers are available for each resource of the link. It is described in the database and managed.

  This makes it possible to manage each link resource.

  In the present invention (Claim 3), the resource allocating means allocates a link resource for a working path or a protection path, and determines whether the working path or the protection path can be used for each resource of the link. It is described in the database and managed.

  According to this, management for each resource of the link can be performed from another viewpoint.

    Furthermore, in the present invention (Claim 4), the resource allocation means allocates link resources for one or more specific customers, arbitrary customers, or backup paths, and which customers can use which link resources. Is described in a database for each resource of the link and managed.

  According to this, it is possible to manage each link resource more finely.

  In the present invention (Claim 5), the policy setting means may specify a resource designated as usable by a specific customer or a resource usable by an arbitrary customer in the resource designation used by the working path and the backup path. It is described in the database that either one or both can be used, and a resource allocation policy is set for each customer based on the contents of the database.

  According to this configuration, a resource allocation policy can be set for each customer.

  In the present invention (Claim 6), the policy setting means describes in a database whether or not a customer allows or disallows sharing of resources with other customers, and sets a resource allocation policy for each customer.

  According to this configuration, it is possible to set a resource allocation policy for each customer in more detail.

  In the present invention (Claim 7), the policy setting means uses at least one resource for the working path from a resource designated as usable by a specific customer and a resource designated as usable by any customer. One or more of the resources designated as available to a specific customer, the resources designated as available to any customer, and the resources made available for the backup path Is used for the backup path in the database, and a resource allocation policy is set for each customer based on the contents of the database.

  According to this configuration, a resource allocation policy can be set for each customer.

  According to the present invention, in a multi-customer accommodating network, not only can efficient resource management be performed based on a customer's request during normal operation, but also recovery can be performed according to a policy related to resource allocation based on a customer's request when a network failure is recovered. It becomes possible.

An embodiment of the present invention will be described in detail with reference to the drawings.
FIG. 1 shows a network configuration premised on the present embodiment. FIG. 1A is a diagram showing an example of the most basic form of a network assumed in the present embodiment, which is composed only of nodes and links. Nodes n1 to n6 have a switching function and a function of setting a path between the nodes. For example, it corresponds to MPLS LSR (Label Switching Router), wavelength cross-connect, and the like. A link is a medium that connects nodes and transfers information. For example, it corresponds to an optical fiber. A path may have a discrete band, such as a wavelength path, or a continuous band, such as an MPLS LSP.

  FIG. 1B is a diagram for supplementarily explaining an example of a basic network configuration assumed in the present embodiment. FIG. 1 (b) is also a configuration example consisting of six nodes from node n1 to node n6 in FIG. 1 (a). In FIG. 1 (b), the directionality of the link is distinguished, and in addition to the current A link prepared for a working path used for operation at the time and a link prepared for a backup path prepared for backup are shown. The 16 links from the link l1 to the link l16 have their respective directions, but the links l5 and 16 shown by broken lines connecting the nodes n1 and n3 indicate links for the backup path. A link l3 and a link l4 indicated by solid lines connecting the same node n1 and node n3 are links for the working path connecting the same section. Note that the symbols from n1 to n6 used in the description here represent the node IDs of these nodes.

  In the following explanation, it is assumed that one customer uses one path (although one path transfers only one customer's traffic), this is not necessary. It doesn't matter. That is, one customer may use one path.

[Configuration of network failure recovery management device]
FIG. 2 is a diagram illustrating a configuration example of the network failure recovery management apparatus 1 in the present embodiment. In this embodiment, link resources (resources) are allocated and managed from two viewpoints. First, the first viewpoint is resource management for each customer. Assign resources to one or more specific customers or any customer. The second point of view is whether the link resource is for the working path or the protection path. By combining these two viewpoints, it is managed which customer can use which resource for each resource of the link. Further, it is managed whether or not the resources for the backup path are allowed to be shared with other customers. The network failure recovery management device 1 may be mounted in a node or in a device different from the node.

  The network failure recovery management apparatus 1 according to the present embodiment includes six management means and four databases. The management means includes resource allocation means 11, policy setting means 12, working path route calculation means 13, backup path route calculation means 14, path setting means 15 and path switching means 16, and the database includes a policy database (hereinafter referred to as a policy database). There are a policy DB and description 21, a path database (hereinafter referred to as path DB) 22, a link resource database (hereinafter referred to as link resource DB) 23, and a spare resource database (hereinafter referred to as spare resource DB) 24.

[Network failure recovery management device database]
The policy DB 21 is a database that stores resource allocation policies for path setting for each customer set by the policy setting unit 12. The policy DB 21 includes items of VPNID, working path resource policy, protection path resource policy, and resource sharing policy.

この表1の例は、図１（ｂ）の例に対応している。また、後記する他のデータベースの内容例とも対応している。 The VPNID is information that uniquely identifies a customer. The working path resource policy designates the range of resources that can be allocated to the working path using an arbitrary combination of VPNID and the protection path resource policy designates the range of resources that can be assigned to the protection path. The resource sharing policy specifies whether or not VPN resources used by other customers can be shared. An example of the contents of the policy DB 21 is shown in Table 1.

The example of Table 1 corresponds to the example of FIG. Moreover, it corresponds also to the example of the content of the other database mentioned later.

  The path DB 22 stores information on paths set by the path setting unit 15. The path DB 22 includes items of path ID, VPNID, bandwidth, route link list, working / reserve flag, and corresponding path ID.

The path ID is information that uniquely identifies the path. The VPNID is the same as the VPNID described above. The bandwidth indicates a bandwidth (resource) used by the path. The route link list represents the route of the path as a list of links through which the route passes. In the example of FIG. 1B, the link ID between node 1 and node 3 is l3, the link ID between node 3 and node 4 is l9, the link ID between node 4 and node 2 is l8, and node 1 to node 2 is established and the route is node 1 -node 3 -node 4 -node 2, the route link list is <l8, l9, l3>.

表２に含まれている経路リンクリストの例は、リンクリソースＤＢ２３の内容例に含まれているリンクＩＤと対応している。 The working / protection flag indicates whether the path to which the link belongs is the working path or the protection path. The corresponding path ID is an item having information indicating the correspondence between the working path and the protection path used when the working path fails. In this item, the path ID of the corresponding protection path is stored in the case of the working path, and the path ID of the corresponding working path is stored in the case of the protection path. An example of the contents of the path DB 22 is shown in Table 2.

The example of the route link list included in Table 2 corresponds to the link ID included in the content example of the link resource DB 23.

  FIG. 3 is a diagram showing the contents of Table 2. For example, the working path VPN1a of VPN1 in Table 2 is represented by a link 17 (refer also to FIG. 1 (b) as appropriate) from the node n2 to the node n4. On the other hand, the backup path VPN1b corresponding to the recovery at the time of failure is a path indicated by a broken line from the node n2 to the node n4 through the nodes n1 and n3. In this route, it passes through links l2, l5 and l9. 3, VPN2 and VPN3 in Table 2 also describe the working path and the protection path as in VPN1, the working path VPN2a of VPN2 corresponds to the protection path VPN2b, and the working path VPN3a of VPN3 similarly. Corresponds to the backup path VPN3b.

  The link resource DB 23 stores the remaining resources of the resources allocated to each customer. As shown in FIG. 2, the link resource DB 23 includes a link ID, an A-end node ID, a Z-end node ID, an A-end IF (Interface) ID, a Z-end IFID, a SRLG (Shared Risk Link Group) ID, an allocation resource, It includes items of remaining resources, VPNID, and working / reserve flags.

  The link ID is information that uniquely identifies the link. The A-end node ID is the node ID of the upstream node of the link, and the Z-end node ID is the node ID of the downstream node of the link.

The IFID is information that uniquely identifies the link (interface to the link) at the node. The A end IFID is the IFID of the link at the A end node, and the Z end IFID is the IFID of the link at the Z end node.
SRLGID assumes a case where multiple links fail for a single failure (for example, when multiple links are accommodated in the same pipeline and a failure occurs in the pipeline) It is information indicating a group of links that can simultaneously fail. A plurality of links having the same value can fail simultaneously with respect to a single failure.

  The VPNID is the same item as the VPNID. In the present embodiment, the link may be assigned to an arbitrary customer, or may be assigned to one or more customers, and either “any” or “one or more lists of VPNIDs” is Is set.

  The allocated resource is an item indicating how many resources, that is, bands are allocated to the link. The remaining resource is an item indicating the remaining resource obtained by subtracting the resource used in the working path and the resource reserved for the backup path from the allocated resource. Note that the resources reserved for the backup path are subject to subtraction only for the minimum necessary bandwidth due to the resource sharing effect. For example, even if there are three backup paths that consume one band, three bands are not required, and three backup paths share a band, and one band is sufficient as a whole.

表３の例は、図１（ｂ）１６本のリンクを含む例であり、このうち、ノードｎ１とノードｎ３の間のリンクには、現用パスのためのリンクｌ３，ｌ４及び予備パスのためのリンクｌ５，ｌ６を含んでいる。 The working / backup flag indicates whether the corresponding resource can be used in the working path, the protection path, or both, and any one of “working”, “reserve”, or “both” is set. Table 3 shows an example of the contents of the link resource DB 23.

The example of Table 3 is an example including 16 links in FIG. 1 (b). Among these, the links between the nodes n1 and n3 are links 13 and 14 for the working path and a backup path. The links l5 and l6 are included.

  In the above configuration, as shown in FIG. 1B and Table 3, the number of links formed is not necessarily one even between the same apparatus and the same IF. When resources are allocated separately for allocation of “working”, “reserve”, and “both” between the same apparatus and the same IF, a plurality of links are formed, and they have different link IDs. However, a configuration in which a single link ID is assigned between the same device and the same IF to form one link, and resources are allocated to each VPNID, backup / working, and both in one link. But it doesn't matter.

  The backup resource DB 24 stores details of information on resources allocated for backup paths. As shown in FIG. 2, the backup resource DB 24 includes a link ID, a common resource item group, and an individual VPN resource item group. In the common resource item group, there are items related to the reservation resource and the SRLGID resource, and the item related to the SRLGID resource further includes two items of the SRLGID and the SRLG reservation resource. The item group of the individual VPN resource includes a VPNID item in addition to the item of the common resource item group. That is, there are items relating to VPNID, reservation resource, and SRLGID resource, and the item relating to this SRLGID resource further includes two items, SRLGID and SRLG reservation resource.

  The link ID is information that uniquely identifies the link. The common resource refers to a resource consumed by a spare path of a customer who may share the resource with other customers. And the item group regarding a common resource stores the information about the backup path which shared the resource in this way. The reserved resource of the common resource is an item indicating a resource reserved for the protection path including a resource sharing effect between the protection paths.

  The SRLGID resource of the common resource is an item indicating a resource that needs to be reserved for the backup path for each SRLGID allocated to the working path. Here, the SRLGID of the SRLGID resource of the common resource indicates the ID of the SRLG of the working path.

  The SRLG reservation resource of the SRLGID resource of the common resource is an item indicating a resource that needs to be reserved in the protection path with respect to the SRLG of the working path.

  There is a reason why the backup resource DB 24 and the backup path route calculation means 14 must manage resources in accordance with the SRLGID of the working path. This is because the risk of multiple failures must be addressed at the same time. Therefore, the corresponding working path and protection path should not be in the same SRLGID group. In addition, a failure may occur simultaneously in a plurality of working paths that belong to the same group with the SRLGID. In this case, what is required as a backup path resource is the total value of the bandwidths of a plurality of working paths that belong to the same group with the SRLGID. For this reason, it is necessary to manage spare resources corresponding to the SRLGID of the working path.

  Unlike the case of the common resource, the individual VPN resource is a resource reserved for a backup path assigned to a customer who cannot share resources with other customers. Even in the case of individual VPN resources, there is a resource sharing effect between backup paths of the same customer, but a sharing effect across a plurality of customers is not assumed.

表４のｌ９のリンクの例でもわかるように、１つのリンクに対してＳＲＬＧＩＤリソースのエントリは必ずしも１つとは限らず、複数のエントリを持つ場合もある。また、リンク全体で予備パス用に予約しておくリソースは、共通リソースの予約リソースと、個別ＶＰＮリソースのそれぞれのＶＰＮの予約リソースの合計となる。これは、予備パス用に用いるリソースにおいて、共通リソースで予約するリソースと個別のカスタマが予約するリソースは、共用できないことを意味している。 The SRLGID resource of the individual VPN resource is an item group indicating a resource that needs to be reserved for the protection path for each SRLGID of the working path, and the SRLGID of the SRLGID resource of the common resource indicates the SRLGID of the working path, and the common resource The SRLG reservation resource of the SRLGID resource indicates a resource that needs to be reserved in the protection path with respect to the SRLGID of the working path.

As can be seen from the example of the link 19 in Table 4, there is not necessarily one SRLGID resource entry for one link, and there may be a plurality of entries. Further, the resources reserved for the protection path for the entire link are the total of the reservation resources of the common resources and the reservation resources of the respective VPNs of the individual VPN resources. This means that in the resources used for the backup path, the resource reserved by the common resource and the resource reserved by an individual customer cannot be shared.

[Means for disaster recovery]
The means for failure recovery in this embodiment performs processing for failure recovery by referring to the four types of databases described above. The resource allocation unit 11 sets the value of each item in the link resource DB 23. The policy setting unit 12 sets each value of the policy DB 21. The working path route calculation means 13 refers to the link resource DB 23 to calculate the working path route and stores the result in the path DB 22. Backup path route calculating unit 14 refers to the link resource DB23 and pre 備Ri source DB 24, and calculates the route for the backup path, and stores the result in the path DB 22. The path setting unit 15 extracts necessary information from the path DB 22 and sets a path. When a failure occurs, the path switching means 16 takes out a backup path route calculated in advance for the working path where the failure has occurred from the path DB 22 and sets and switches the path along this path. The movement of each means will be described through specific examples of processing to be described later.

[Operation of network failure recovery management device]
The operation of the network failure recovery management device 1 will be described through the description of the operation of each means for recovery. First, the operation of the working path route calculation unit 13 will be described.

FIG. 4 is a diagram for explaining a method in which the working path route calculation unit 13 sets the working path.
The working path route calculation means 13 first extracts a link that matches the working path policy (S11). At this time, specifically, processing is performed corresponding to the following three cases corresponding to the working path policy.

  When the working path policy of the corresponding customer is “resources designated as being usable by a specific customer”, the link assigned to the corresponding customer is extracted.

  If the working path policy of the corresponding customer is “a resource designated to be available to an arbitrary customer”, a link assigned to the arbitrary customer is extracted.

  When the working path policy of the corresponding customer is “resources specified to be usable by a specific or arbitrary customer”, the link assigned to the corresponding customer or the arbitrary customer is extracted.

  Next, a link assigned to the working path is further extracted from the links extracted in step S11 (S12). Specifically, links are extracted according to the following three cases.

  If a link is a “link allocated for the working path”, the link is extracted.

  If a link is a “link allocated for a protection path”, the link is not extracted.

  If a link is “a link assigned to both the working path and the protection path”, the link is extracted.

  Further, referring to the link resource DB 23, a link whose residual resource is equal to or larger than the requested bandwidth of the path is extracted from the extracted links (S13).

  Then, a link is selected from the links extracted in the steps so far by a known method, and the shortest path calculation is performed (S14). If no route is found in this calculation, a route calculation error occurs and the process ends. An example of a specific calculation method for this step will be described later with reference to FIG.

  Then, receiving the calculation result of step S14, the contents of the path DB 22 are updated (S15), and the link resource DB 23 is also updated (S16). When the path DB 22 is updated, a path ID is assigned, a VPNID, a bandwidth, and a link list of path calculation results are stored, and the current / backup flag is set to “current”. In response to this calculation result, the path setting means 15 actually sets a path.

  FIG. 5 is a diagram for explaining an example of the shortest path calculation performed in step S14 of the working path setting method, that is, the network shortest path search method.

  First, a network node and link distance (cost) are prepared so that they can be referred to (S21). In this embodiment, since all links are treated as equidistant, there is no particular item in the database. However, in an embodiment that handles links with different distances, the link resource DB 23 stores the link distance (cost). ) Item is provided and the value is saved.

  Next, all the nodes other than the starting point are stored in the group of unexamined nodes, and the starting point node is put in the group of already checked nodes to perform initialization (S22).

  Then, the shortest link coming out from the node of the investigated group is searched, and the shortest node is selected from the starting point in the uninvestigated group (S23).

  Then, it is checked whether or not the selected node is an unchecked one (S24). As a result, if it has not been investigated (Yes in S24), the node is added to the investigated group, the shortest distance and route to the node are stored (S25), and the process proceeds to step S26. If it has already been investigated (No in S24), the process proceeds to step S26 without doing anything.

  At that time, a group of unexamined nodes is examined to determine whether there are any unexamined nodes remaining (S26). When it remains (Yes of S26), it returns to S23 and repeats a search. If there is no longer left (No in S26), the shortest route has been calculated, so the shortest distance and route of the node at the arrival point are output (S27), and the process ends.

  FIG. 6 is a diagram showing a procedure for executing the path calculation of the backup path by the backup path path calculation unit 14. In this method, a link is extracted from each of the link resource DB 23 and the spare resource DB 24, and a route is calculated from a group of links obtained by combining these links.

  First, the link resource DB 23 is referred to, and the link having the SRLGID that matches at least one of the SRLGIDs of the links used by the working path is excluded (S31).

  Next, the link resource DB 23 is referred to, and a link that matches the protection path policy is extracted (S32). Specifically, the process is performed in the following three cases.

  When the backup path policy of the relevant customer is “resources designated as being usable by a specific customer”, the link assigned to the relevant customer is extracted.

  When the backup path policy of the corresponding customer is “a resource designated as being available for an arbitrary customer”, a link assigned to the arbitrary customer is extracted.

  When the backup path policy of the corresponding customer is “resources specified to be usable by a specific or arbitrary customer”, the link assigned to the corresponding customer or the arbitrary customer is extracted.

  Next, links that can be used for backup are extracted from the links (S33). In that case, specifically, the process is divided into three cases as follows.

  If a link is “a link allocated for the working path”, the link is not extracted.

  If a link is a “link allocated for a protection path”, the link is extracted.

  If a link is a “link assigned to both the working and backup paths”, the link is extracted.

  And the value of the residual resource of the link extracted in the steps so far is obtained from the link resource DB 23 and stored together with the link (S34). Up to this point, extraction of the link from the link resource DB 23 ends.

  Next, it is checked whether or not there is a link entry that can be used for backup by referring to the backup resource DB 24 (S35).

  If there is no such entry (No in S35), the extraction of the link from the spare resource DB 24 is terminated, and the route calculation of S39 is prepared with the link from the link resource DB 23. When there is an entry for a link that can be used for backup (Yes in S35), a link having an SRLGID that matches at least one of the SRLGIDs of the link used by the working path is excluded from the entry (S36).

  Further, a link that matches the resource sharing policy is extracted (S37). Specifically, the link that has an entry in the spare resource DB 24 is processed in the following two cases.

  When the resource sharing policy of the corresponding customer is “Do not allow sharing with other customers”, the link having the entry of the VPNID corresponding to the individual VPN resource is extracted.

  If the resource sharing policy of the corresponding customer is “permit resource sharing with other customers”, a link having an entry in the common resource is extracted.

  Then, the minimum reserved resource vacancy is obtained and stored in the SRLGs of all the working paths (S38). Specifically, the minimum value of reserved resources is obtained by the following procedure. This method is slightly different depending on the resource sharing policy. Details of how to obtain the minimum value will be described later with reference to FIG. 7, and only an outline will be described here.

  First, in the case where sharing with other customers is not permitted, when there is no entry corresponding to the SRLG of the working path, the reserved resource of the individual VPN resource corresponding to the relevant VPNID of the relevant link is saved, and the working path When there is an entry corresponding to the SRLG, a value obtained by subtracting the SRLG reservation resource of the SRLDID from the reservation resource of the corresponding VPNID resource of the corresponding link is stored. Then, these processes are checked for all SRLGIDs of the working path, and the minimum value is stored.

  When sharing with other customers is allowed, when there is no entry corresponding to the SRLG of the working path, the reserved resource of the common resource of the corresponding link is saved, and when there is an entry corresponding to the SRLG of the working path The value obtained by subtracting the SRLG reservation resource of SRLDID from the reservation resource of the common resource of the corresponding link is stored. Then, these processes are checked for all SRLGIDs of the working path, and the minimum value is stored.

  Next, by using the remaining resources obtained from the link resource DB 23 and the reserved resources available from the spare resource DB 24, a link usable for the spare path is obtained and stored (S39). At this time, the information extracted by S34 and the link extracted by S38 are also prepared for use in route calculation.

  There are two methods for calculating in the step of S39. First, using the information stored in S34, the first method tries to calculate a route after extracting a link exceeding the requested bandwidth from this link. It is a method of using the information. The information from S38 includes the minimum value of the link and the resource, but the information of the link and the remaining resource of S34 is supplemented with the minimum value of the resource of S38 for each link to supplement the bandwidth. In this method, the bandwidth that can be used for each link is obtained, and a link that satisfies the requested bandwidth is extracted from this and used for route calculation.

  On the other hand, the second method supplements the bandwidth by adding the minimum value of the resource of S38 for each link to the information of the link of S34 and its remaining resources from the beginning, and obtains the bandwidth that can be used for each link. In addition, a link that satisfies the required bandwidth is extracted from these and used for route calculation.

  The shortest path is calculated from the links thus given using the example of the calculation method described with reference to FIG. 5 (S40). Then, the path DB 22 is updated based on the calculation result (S41), the backup path is added, and the link resource DB 23 and the backup resource DB 24 are also updated together (S42). The procedure for updating the link resource DB 23 and the spare resource DB 24 in step S42 will be described in detail with reference to FIG.

  FIG. 7 is a diagram for explaining a method for obtaining the minimum value of free resources in step S38 of FIG. First, one of the SRLGIDs of the working path is selected. Then, the initial value of the variable MIN for storing the minimum value is set to a large value and initialized (S51).

  Next, it is checked whether or not the SRLGID entry exists in the entry of the corresponding link in the spare resource DB (S52).

  As a result of the examination, if there is no entry of the SRLGID selected first (No in S52), the value of the reserved resource is extracted according to the resource sharing policy. First, in the case of a policy that does not allow resource sharing, the value of the reserved resource of the individual VPN resource corresponding to the VPNID of the customer whose entry is applicable is extracted. If resource sharing is permitted, the value of the reserved resource of the common resource is extracted (S53). This value is used in step S55 for comparison with MIN.

  If there is an entry of the SRLGID selected first (Yes in S52), the amount of resources that can actually be used is determined according to the resource sharing policy. First, in the case of a policy that does not allow resource sharing, the amount of resources that can be used by subtracting the value of the SRLG reserved resource of the corresponding SRLGID from the reserved resource of the individual VPN resource corresponding to the VPNID of the customer whose entry is the corresponding Calculate When resource sharing is permitted, the amount of usable resources is calculated by subtracting the value of the SRLG reserved resource of the corresponding SRLGID from the reserved resource of the common resource (S54). This value is used in step S55 for comparison with the next MIN.

  Then, the value obtained in S53 or S54 is compared with the value of MIN stored at that time, and it is checked whether or not the value is smaller than MIN (S55).

  When the value is small (Yes in S55), the value of MIN is updated to the above value and set as a new minimum value (S56). Then, the process proceeds to step S57. If the value is greater than or equal to MIN (No in S55), the process proceeds to step 57 as it is.

  In the process so far, the processing for one SRLGID is completed. At this point, it is checked whether or not all SRLGIDs of the working path have been confirmed (S57). If it has been completed (Yes in S57), the value of the MIN at that time is stored as the minimum value of available resources (S58), and the process is terminated. If another SRLGID still remains (No in S57), the next SRLGID of the working path is selected (S59), and the process returns to step S52 to repeat the process.

FIG. 8 is a diagram for explaining the update of the link resource DB 23 and the spare resource DB 24 in the last step S42 of FIG. First, the backup path calculation result is received, and it is checked whether there is an entry for the link to be updated in the backup resource DB 24 (S61).

  When there is an entry to be updated (Yes in S61), it is checked whether there is an entry corresponding to the resource sharing policy (S63). If there is no such entry (No in S63), the remaining resource is reduced by the bandwidth of the newly set backup path (S62), and the processing after Step S64 is continued. If there is a target entry (Yes in S63), the process proceeds to step S65 described later, and the spare resource DB 24 is updated.

  If there is no entry for the link to be updated (No in S61), for the update target link in the link resource DB 23, first, the remaining resource is reduced by the bandwidth of the newly set backup path (S62). Then, the process proceeds to step S64.

  In step S64, it is determined in which table an entry should be created for a link that does not yet have an entry in the backup resource DB 24. Specifically, it is checked whether or not the resource sharing policy of the link allows sharing of resources with other customers (S64).

  When the resource sharing policy permits sharing (Yes in S64), first, an entry is added to the common resource table of the spare resource DB 24 (S69). Then, an entry is created for each SRLGID related to the common resource table (S70). Next, the used bandwidth of the protection path to be newly set is entered as the SRLG reserved resource of the SRLGID resource (S71). Similarly, the used bandwidth of the protection path is entered as the reserved resource of the common resource (S72), The process ends.

  If the resource sharing policy does not permit sharing (No in S64), first, an entry is added to the individual VPN resource table of the backup resource DB 24 (S73). Then, an entry is created for each SRLGID in the individual VPN resource table managed for each customer (S74). Next, the used bandwidth of the protection path is entered as the SRLG reserved resource of the SRLGID resource (S75), the used bandwidth of the protection path is entered as the reserved resource of the individual VPN resource (S76), and the update process is terminated.

  In step S65, the SRLG reservation resource value of the SRLGID of the path to be updated is increased by the bandwidth used for the backup path (S65). Then, it is checked whether or not the value of the SRLG reservation resource increased here exceeds the existing reservation resource (S66). This is to check whether there is a link that cannot cover the bandwidth of the newly set protection path only by using the resource reserved for the protection path. If there is no such shortage of reserved resources (No in S66), there is no need for further updating, and the updating process is terminated. On the other hand, if there is a link with insufficient reserved resources (Yes in S66), the reserved resources are increased by the shortage for the backup path in the backup resource DB 24 (S67), and the remaining resources of the corresponding link are stored in the link resource DB 23. The deficiency is reduced (S68), and the update process is terminated.

[Embodiments based on other database configurations]
FIG. 9 is a diagram for explaining an example of another embodiment of the network failure recovery management apparatus 1. Comparing FIG. 9 with FIG. 2 showing the embodiment described so far, only the items managed by the link resource DB 23 are different, and the other configurations are all the same. Therefore, only the parts that are handled differently will be described here.

  The most different point in this embodiment is that there is no longer an active / reserve flag item in the link resource DB 23. As a result, in link level management, the hardware of one link is assigned to the working path in order not to distinguish the link assigned to the working path from the link assigned to the protection path. It is no longer necessary to divide and manage those allocated to backup paths.

  In the present embodiment, management of the backup path policy in the policy DB 21 is changed instead of distinguishing between the current path and the backup path in the link resource DB 23. Specifically, the backup path policy includes “resources designated for use by backup path” in addition to “resources designated for use by a specific customer” and “resources designated for use by any customer”. It is possible to specify that any one or more of these three can be used. Furthermore, “arbitrary”, “one or more lists of VPNIDs” or “reserve” are set as the VPNID of the link resource DB 23.

  Along with this, a slight change occurs in the working path route calculation unit 13 and the backup path route calculation unit 14. First, in the working path route calculation means 13, since there is no longer any distinction between the working path and the protection path, the step S12 in FIG. 4 is eliminated, and the working link is not extracted. In the backup path route calculating means 14, since the resource policy has been changed, the first step of extracting the link corresponding to the policy is changed. Specifically, instead of step S32 and step S33 in FIG. 6, the following step of extracting one link is performed.

  When the backup path policy of a customer who is assigned a new backup path includes “resources designated as being available for a specific customer”, the link assigned to the customer is extracted. When the backup path policy of the customer includes “a resource designated to be usable by an arbitrary customer”, a link assigned to the arbitrary customer is extracted. When the backup path policy of the customer includes “resources designated for the backup path”, the link allocated for the backup path is extracted.

  In the present embodiment, the database management is slightly simplified. However, no particular inconvenience arises in network failure recovery. Then, the management of the link resource DB 23 that describes the resource of the link can be easily performed.

  As described above, in both of the two embodiments, it is possible to finely allocate link resources for each of the working path and the protection path based on the customer policy. As a result, even in the event of a network failure, the network can be quickly restored in a manner that meets the customer's request while minimizing the impact on the customer's network operation. Furthermore, risk management can be performed to minimize the risk of network failure before the occurrence of network failure. In either of the two embodiments, the resource allocation unit 11, the policy setting unit 12, the working path route calculation unit 13, the protection path route calculation unit 14, the path setting unit 15 and the path switching unit 16 This is realized by executing a program by an arithmetic device provided in the recovery management device.

(A) is a figure which shows the example of the most basic form of the network assumed in this embodiment, (b) is a figure which shows the example of the form which considers the directionality of a link. 2 is a diagram illustrating a configuration example of a network failure recovery management device 1. FIG. It is a figure which shows the example of a path | route. It is a figure which shows the procedure which performs the path | route calculation of a working path by a working path path | route calculation means. It is a figure explaining an example of the network shortest route search method. It is a figure which shows the procedure which performs the path | route calculation of a backup path by a backup path route calculation means. It is a figure explaining the method of calculating | requiring the minimum value of the free resource in step S38 of FIG. It is a figure explaining the operation | movement of update of link resource DB23 and backup resource DB24 in step S42 of FIG. FIG. 9 is a diagram for explaining an example of another embodiment of the network failure recovery management apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Network failure recovery management apparatus 11 Resource allocation means 12 Policy setting means 13 Working path route calculation means 14 Backup path route calculation means 15 Path setting means 16 Path switching means 21 Policy database (policy DB)
22 Path database (path DB)
23 Link Resource Database (Link Resource DB)
24 Spare resource database (spare resource DB)

Claims (8)

In a multiple customer accommodation network that accommodates multiple customers in the same network,
Network failure recovery management device
A resource allocating unit having a function of setting to which customer the resource of each link is allocated, and a function of setting whether to allocate the link to the working path or the protection path;
A policy setting means for setting a resource allocation policy for each customer;
A working path route calculating means for calculating a route of a working path using resources available to each customer;
A backup path calculation means for calculating a backup path using resources available to each customer;
Path setting means for setting a path calculated by the working path route calculating means or the backup path route calculating means;
Path switching means for switching paths when a network failure occurs,
The resource allocation means allocates the resource to a specific customer or shared resource, and then allocates the resource to a working path or a protection path, respectively;
The policy setting means sets a resource allocation policy for each customer;
The working path route calculating means calculates a working path,
The backup path route calculating means calculates a backup path,
The path setting means sets the working path and the protection path in a network;
A network failure recovery management method, characterized in that, when a network failure occurs in the working path, the path switching means switches a path to the preset backup path.
The resource allocation means allocates link resources to one or more specific customers or any customer;
The network failure recovery management method according to claim 1, wherein which customer is available is described and managed in a database for each resource of the link. The resource allocation means allocates link resources for a working path or a protection path;
3. The network failure recovery management method according to claim 2, wherein whether the active path or the backup path can be used is described in a database for each link resource and managed. The resource allocating means allocates link resources for one or more specific customers, any customer, or backup path;
2. The network failure recovery management method according to claim 1, wherein which customer can use which link resource is described in a database for each link resource and managed. In the resource specification used by the policy setting means used by the working path and the protection path,
Describe in the database that either one or both of the resources designated as available to a particular customer or the resources available to any customer can be used;
The network failure recovery management method according to any one of claims 1 to 4, wherein a resource allocation policy is set for each customer based on the contents of the database. The policy setting means is
Describe in the database whether the customer allows or disallows sharing of resources with other customers,
6. The network failure recovery management method according to claim 5, wherein a resource allocation policy is set for each customer. The policy setting means is
That at least one resource is available for the working path from a resource designated as available for a particular customer and a resource designated as available for any customer;
One or more resources can be used for the protection path from the resources designated as usable by a specific customer, the resources designated as usable by any customer, and the resources made available for the protection path. Describe that in the database,
The network failure recovery management method according to any one of claims 1 to 4, wherein a resource allocation policy is set for each customer based on the contents of the database. A network failure recovery management apparatus that executes the network failure recovery management method according to any one of claims 1 to 7.

Images