JP4448474B2 - Node equipment and node equipment - Google Patents

Description

  The present invention relates to a node device and a node device that collect network information and set a path between nodes across a plurality of domains in a network in which a plurality of domains including nodes and links operated according to a predetermined policy exist.

  A plurality of domains exist in the network, and node addresses (hereinafter referred to as node addresses) may overlap between the domains. In such a case, in order to exchange information about a link connecting adjacent nodes (hereinafter referred to as link information) between nodes across domains, conventionally, link information (network information) of one domain is exchanged with another domain. There is a method of not passing to the network, or a method of converting a node address when link information of a certain domain is passed to another domain (see Patent Document 1).

In these methods, even if the node address for uniquely identifying the link is used redundantly between domains, it is not advertised to another domain, or the uniqueness of the node address is maintained by address translation. Therefore, it is not necessary to perform cooperative address assignment in each domain, and a private address (user-set address) can be given.
International Publication No. WO 2004/068805 A1 Pamphlet

However, with the conventional method of not passing link information of one domain to another domain, link information of another domain cannot be used when performing route calculation to set a path that crosses multiple domains. Therefore, it is difficult to obtain an optimal path calculation result.
In addition, in the method of converting a node address when passing link information of a certain domain to another domain, it is necessary to hold and manage an address conversion table.

  In view of the above problems, an object of the present invention is to provide means for easily collecting network information and setting a path even if node addresses are duplicated among a plurality of domains in a network. .

The present invention for solving the above problems, is operated in accordance with a predetermined policy, the nodes and the network domain there are a plurality including a link connecting between the nodes, Ri span multiple domains, unique within the domain A node device that collects information about a network and sets a path between nodes to which an address is assigned, and includes link information storage means for storing link information that is information on each link in the network, and link information storage means the link information stored, and a domain identifier assigning means for assigning a unique domain identifier in the domain where the link belongs, and address information indicating an address within the domain of the node domain identifier assigned by the domain identifier providing means Combined to uniquely identify the entire network And link information exchanging means for the link information exchanging with other nodes, using the link information, a route calculation means for calculating a route path between a given node in the network, according to the calculated route by the route calculating means And path setting means for transmitting / receiving path setting information including a path between nodes constituting the path, and holding the path setting information.

  The present invention is also a node device, domain identifier filtering information storage means for storing domain identifier filtering information indicating a domain identifier of link information to be transmitted from the node device to each adjacent node, and domain identifier filtering The system further comprises domain identifier filtering means for extracting corresponding link information from the link information storage means in accordance with the domain identifier filtering information stored in the information storage means.

  In addition, the present invention provides a node device, a group identifier giving means for assigning a group identifier specific to a group to which the link belongs to link information stored in the link information storage means, and the node device to each adjacent node. Corresponding from the link information storage means according to the group identifier filtering information storage means for storing the group identifier filtering information indicating the group identifier of the link information to be transmitted to the group identifier filtering information stored in the group identifier filtering information storage means It further comprises group identifier filtering means for extracting link information.

  Also, the present invention provides a node device having a plurality of node devices, which is a communication path for exchanging link information and path setting information between node devices in the node device and between node devices across the node devices. A link is provided.

  The link information storage means, the domain identifier filtering information storage means, and the group identifier filtering information storage means in the claims are respectively provided in the link DB 32, the domain identifier filtering DB 34, and the group identifier filtering DB 35 in the best mode for carrying out the invention. Correspond.

  According to the present invention, even if node addresses overlap between a plurality of domains in a network, link information of a certain domain can be easily passed to another domain. This makes it easy to collect network information and set an optimal path between nodes that span multiple domains.

  Further, according to the present invention, link information exchanged between nodes can be filtered by a domain to which the node belongs or an arbitrary group. As a result, when setting a path between nodes, it is possible to pass or not pass nodes belonging to a given domain or group, thus improving security, maintenance and operability. Can do.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings.

≪Network configuration and overview≫
FIG. 1 is a diagram showing a network configuration according to an embodiment of the present invention. The network is configured by connecting a node, a BN (Border Node), and a link for each domain. Nodes (node devices) in the domain and links connecting the nodes are operated according to a predetermined policy. The domains are connected by BN.

  In FIG. 1A, links are set between nodes in each domain. For example, in domain # 2, link c is set between node # 4 and node # 5. BN means a node connected to a node belonging to another domain among nodes belonging to a predetermined domain. For example, between domain # 1 and domain # 2, BN # 1 and BN # 5 are connected via a link g. It should be noted that a link does not necessarily exist between a plurality of nodes included in the same domain, such as between node # 8 and BN # 9 and BN # 10 and node # 11 shown in domain # 4. Also good. In the following description, BN may be simply referred to as a node.

  As shown to Fig.1 (a), the some BN of a different domain exists in a physically different apparatus, and between BN (between apparatuses) may be connected by the link. In this case, the link between BNs belongs to both domains. For example, the link g between BN # 1 and BN # 5 belongs to both domain # 1 and domain # 2.

  Further, as shown in FIG. 1B, a plurality of BNs of different domains may be physically in the same device, and the BNs may be connected within the device. For example, BN # 1 in FIG. 1B represents one device, and the device includes a BN belonging to domain # 1, a BN belonging to domain # 2, and a BN belonging to domain # 4. I have. Details will be described later (see FIG. 9).

  In order to exchange control information such as link information (network information) and path setting information between nodes, a link for transmitting and receiving data may also be used, or a dedicated communication path independent of the link is provided. Also good. Here, a dedicated communication path for exchanging control information is called a control link. The link may or may not be directed. Further, “# 1”, “a”, and the like in FIG. 1 are uniquely assigned in the illustrated network, but for convenience of explanation, they are not included in an actual DB (described later).

<< First Embodiment >>
FIG. 2 is a diagram showing a configuration of the node according to the first exemplary embodiment of the present invention. The node (node device) 2 includes a domain identifier assigning means 21, a link information exchanging means 22, a route calculating means 23, a path setting means 24, an adjacent link DB 31, a link DB (link information storage means) 32, and a path DB 33. Is done. The node 2 is realized by a node device having an exchange capability such as a router or a PC (Personal Computer). The domain identifier assigning means 21, the link information exchanging means 22, the route calculating means 23, and the path setting means 24 are realized by a CPU (Central Processing Unit) built in the node 2 executing a program stored in a predetermined memory. Is done. The adjacent link DB 31, the link DB 32, and the path DB 33 are constructed in a nonvolatile storage device such as a hard disk device that is built in or externally connected to the node 2.

  The adjacent link DB 31 is a DB that stores link information of a link directly managed by the node 2, that is, a link having the node 2 as one end. The stored link information includes an A end node address, a Z end node address, an A end IFID, a Z end IFID, a domain identifier, and a remaining bandwidth. Other information may be included.

  The A-end node address is address information of a node (A-end node) that is one end of the link. The Z end node address is address information of a node (Z end node) which is the other end of the link. The A-end IFID (Interface Identification) is information that uniquely indicates the link in the A-end node. The Z end IFID is information that uniquely indicates the link in the Z end node. The domain identifier is information that uniquely indicates the domain to which the link belongs. The remaining bandwidth is information indicating the bandwidth (data transfer amount [GB / sec] per unit time) remaining in the link (newly assignable).

  For example, an IP address corresponds to the A-end node address and the Z-end node address. Further, the A-terminal IFID and the Z-terminal IFID correspond to, for example, an IP address or a unique number when viewed from the node 2. It is possible to set the address of the node 2 as the A-end node address and set the address of a node adjacent to the node 2 (hereinafter referred to as an adjacent node) as the Z-end node address. It can be set so that the link can be identified without any problem.

  The link DB 32 is a DB that stores the link information included in the adjacent link DB 31 and the link information acquired by the link information exchanging means 22. In other words, the link DB 32 is a DB that stores link information of the entire network. The details of the link information are the same as those of the adjacent link DB 31.

  Here, FIG. 3 is a diagram illustrating an example of a link DB based on the network configuration of FIG. It also describes the assumed node address. The node address is assigned uniquely within the domain. For links with multiple domain identifiers, the first (first) domain identifier means the domain identifier as seen from the A-end node address, and the last (next) domain identifier is seen from the Z-end node address. Means a domain identifier. For example, in the case of the link g, the node # 1 corresponding to the A-end node address “192.168.1.1” belongs to “domain # 1”. Also, the node # 5 corresponding to the Z-end node address “192.168.2.2” belongs to “domain # 2”. Therefore, “1, 2” is set in the domain identifier of the link g.

  Returning to FIG. 2, the path DB 33 is a DB for storing path information between nodes. The path information includes a path ID, an Src address, an Src domain identifier, a Dst address, a Dst domain identifier, a route, and a bandwidth. Other information may be included.

  The path ID is information that uniquely identifies the path. A path is a route between predetermined nodes and passes through one or more links. The path may be uniquely identified only by the path ID, or the path may be uniquely identified by combining with a pair of Src address and Dst address. The Src address is information indicating the address of the node that is the data transmission source of the path. The Src domain identifier is an identifier of the domain to which the path data transmission source node belongs. The Dst address is information indicating the address of the data transmission destination node of the path. The Dst domain identifier is an identifier of the domain to which the node that is the data transmission destination of the path belongs. The route is information indicating a list of links through which the path passes. The bandwidth is information indicating the bandwidth of the path (data transfer amount per unit time [GB / sec]).

  Here, as a list (route) of links through which a path passes, node address information and domain information are combined so that links can be uniquely identified in the entire network. For example, in the example of FIG. 1A and FIG. 3, when the path of the path from the node # 4 to the node # 7 is indicated by a link, when the path information is c−ga−b−h−d, <192.168.1.1,1,2>-<192.168.2.2,2,2>-<192.168.1.1,1,1>-<192. 168.2.2,2,1>-<192.168.3.3,2,1>-<192.168.1.1,1,3>.

  Here, <x, y, z> means x = node address, y = link IFID in a node whose node address is x, and z = identifier of the domain to which the link belongs. For example, in the case of link c, since x = node address of node # 4, y = A-end IFID of link c, and z = domain identifier of link c, <192.168.1.1,1,2> Can be specified. For link b, x = node address of node # 2, y = A-end IFID of link b, and z = domain identifier of link b, so that <192.168.2.2, 2, 1> Can be specified.

  The domain identifier assigning means 21 sets a domain identifier unique to the domain in the link information of the adjacent link DB 31. The trigger for this setting includes an instruction from the operator. Note that methods for setting other information in the link information of the adjacent link DB 31 include setting by an operator and automatic setting for exchanging a part of link information with an adjacent node. This automatic setting between adjacent nodes is to send and receive “LinkSummary Message”, “LinkSummaryAck Message”, etc. according to LMP (Link Management Protocol) studied by the Internet Engineering Task Force (IETF) Is done by.

  The link information exchanging means 22 exchanges information of the link DB 32 with the adjacent node so that the contents of the link DB 32 are the same. The adjacent nodes whose link information is exchanged by the link information exchanging means 22 may be all the adjacent nodes connected by the links included in the adjacent link DB 31 or a new node that specifies the adjacent nodes for exchanging link information. An adjacent DB may be prepared and adjacent nodes included in the DB may be used. In the initial state, the contents of the link DB 32 are the same as those of the adjacent link DB 31, but as the link information is exchanged with the adjacent nodes, the link information of the entire network is reflected. As the link information exchanging means 22, for example, a method defined in RFC (Request for Comments) 3630 (Traffic Engineering (TE) Extensions to OSPF Version 2) can be used. Note that RFC is a document of technical proposals and comments published by IETF.

  The route calculation means 23 uses the link information of the link DB 32 to calculate a path route between designated nodes, and stores the result in the path DB 33.

  The path setting unit 24 communicates with the path setting unit 24 of the adjacent node to set a path. In the first node of the path, the path setting means 24 is calculated by the route calculation means 23, and in order to set the path according to the path (with domain identifier) stored in the path DB 33, the path setting means 24 configures the next node (configures the path Path setting information is transmitted to the node that performs the processing. Here, the path setting information is obtained by adding information such as a band to the route calculated by the route calculation means 23 (including the route). Then, for the link used for path setting (the link that transmitted the path setting information), the path band is subtracted from the value of the remaining band of the link information in the adjacent link DB 31. In order to use the link for path setting, it is necessary to secure a bandwidth necessary for data transfer in the path in the bandwidth of the link, and the remaining bandwidth is updated for that purpose (the same applies hereinafter). ).

  In a node in the middle of the path, the path setting unit 24 receives the path setting information from the path setting unit 24 of another node and adds (holds) it as new information to the path DB 33. Then, the path setting information is transmitted to the next node of the path with reference to the path included in the path setting information. Further, for the links used for path setting (the link that has received the path setting information and the link that has been transmitted), the path bandwidth is subtracted from the value of the remaining bandwidth of the link information in the adjacent link DB 31.

  In the node at the end of the path, when the path setting unit 24 receives the path setting information, it adds (holds) it as new information to the path DB 33. Then, for the link used for path setting (the link that has received the path setting information), the path band is subtracted from the value of the remaining band of the link information in the adjacent link DB 31.

  In each node, when the remaining bandwidth of the adjacent link DB 31 changes, it is reflected in the link DB 32, and the link information is exchanged with the adjacent node so that the link information is the same as the latest. In order to confirm the path, confirmation information corresponding to the path setting information may be transmitted from the node at the end of the path to the node at the beginning of the path via the intermediate node. For such path setting, for example, a method defined in RFC3473 (Generalized Multi-Protocol Label Switching (GMPLS) Signaling Resource ReserVation Protocol-Traffic Engineering (RSVP-TE) Extensions) can be used. Further, the path can be deleted by a similar process.

  In the adjacent link DB 31 and link DB 32 shown in FIG. 2, the domain identifier is set for each link, but one domain identifier is associated with the A-end node address, and another Z-node address is assigned. You may set so that a domain identifier may be matched.

  FIG. 4 is a flowchart showing processing for setting a path by a node. First, in the node 2, the domain identifier assigning means 21 assigns a domain identifier to the link information of the adjacent link DB 31 (step S401). This is due to operator activation.

  Next, the link information exchanging means 22 exchanges the link information of the link DB 32 with an adjacent node (step S402). This is due to the activation of the operator or the activation by updating the link information. As a result, link information of the entire network is acquired.

  Here, it is assumed that an opportunity for path setting has occurred (step S403). The trigger may be an instruction from an operator, an increase in traffic, or the like. As an input in this case, an Src address, an Src domain identifier, a Dst address, a Dst domain identifier, a bandwidth, and the like are given.

  Then, the route calculation means 23 calculates the route of the path (step S404). As this algorithm, the Dijkstra method or the like can be considered.

  Further, the path setting unit 24 sets a path along the calculated route (step S405). When the bandwidth of the link is changed with this path setting, the link information exchanging means 22 notifies the change to the entire network.

  According to the first embodiment described above, a domain identifier is newly defined in addition to the node address as information for uniquely identifying a link. The node address is assigned uniquely within the domain. According to this, the link can be uniquely identified in the entire network by combining the node address and the domain identifier. Further, link information of a certain domain can be passed to another domain without using an address conversion table.

  Further, after calculating the path of the path using this link information, the path is specified including the domain identifier to set the path. As a result, the path of the path can be uniquely identified at a node in the middle of the path. In addition, even when there are overlapping node addresses among a plurality of domains, it is possible to uniquely set or delete a path.

<< Second Embodiment >>
FIG. 5 is a diagram showing a configuration of a node according to the second exemplary embodiment of the present invention. The node 2a includes a domain identifier adding unit 21, a link information exchanging unit 22, a route calculating unit 23, a path setting unit 24, a domain identifier filtering unit 25, an adjacent link DB 31, a link DB 32, a path DB 33, and a domain identifier filtering DB (domain identifier filtering DB). Information storage means) 34. The node 2a has a configuration in which a domain identifier filtering unit 25 and a domain identifier filtering DB 34 are added to the node 2 according to the first embodiment shown in FIG.

  Similar to the domain identifier assigning means 21, the link information exchanging means 22, the route calculating means 23, and the path setting means 24, the domain identifier filtering means 25 executes a program stored in a predetermined memory by the CPU built in the node 2a. It is realized by doing. Similar to the adjacent link DB 31, link DB 32, and path DB 33, the domain identifier filtering DB 34 is constructed in a non-volatile storage device such as a hard disk device built in or externally connected to the node 2a.

  The domain identifier filtering DB 34 is a DB that stores information indicating which domain identifier is assigned to link information from the node 2a to each adjacent node. The stored information (domain identifier filtering information) includes an adjacent node address, a domain identifier, and a target domain identifier for each adjacent node. Other information may be included. Moreover, the setting of these information is performed by an operator's operation.

  The adjacent node address is address information of the adjacent node. The domain identifier is an identifier of a domain to which an adjacent node belongs. The target domain identifier means a domain identifier of link information to be filtered and transmitted to the adjacent node. In other words, link information other than the target domain identifier is not transmitted to the adjacent node. This makes it possible to control whether or not link information belonging to a predetermined domain is transmitted to a specific adjacent node in the node.

  The domain identifier filtering unit 25 filters (extracts) link information in the link information DB 32 using the domain identifier filtering DB 34. Thereby, the adjacent node and the link information exchanging means 22 can selectively exchange link information. Other means and DB are the same as those in the first embodiment.

  The path setting process in this case is different from the process shown in FIG. 4 in that, before the link information exchanging means 22 exchanges link information with an adjacent node in step S402, the domain identifier filtering means 25 obtains the link information of the link DB 32. This is a process that adds filtering.

  FIG. 6 is a diagram illustrating an example in which filtering is performed by the node 2a illustrated in FIG. 5 in the network configurations of FIG. 1A and FIG. FIG. 6A shows setting information of the domain identifier filtering DB 34 in each node, which is a prerequisite. For example, in the domain identifier filtering DB 34, the setting information at node # 1 (information on adjacent nodes) is information on node # 2, node # 5, and node # 9 from the top. FIG. 6B shows link information of the link DB after filtering in each node. Although FIG. 6A shows the setting information in the BN, there may be a domain identifier filtering DB 34 in a node other than the BN.

  Here, in FIG. 6, for a link with one domain identifier, link information corresponding to the target domain identifier is transmitted to the adjacent node. For example, as shown in the setting information in node # 1 in FIG. 6A, node # 1 has a domain identifier of # 1, # 2, # 3, or # with respect to node # 2 (first from the top). 4 link information is transmitted. In other words, the node # 1 does not transmit link information having a domain identifier other than that to the node # 2.

  In addition, for links to which a plurality of domain identifiers are assigned, a rule that excludes links to which domain identifiers other than the domain identifier specified as the target domain identifier are assigned in the domain identifier filtering DB 34 (does not transmit link information) Filter based on As shown in the setting information at the node # 1 in FIG. 6A, the node # 1 transmits link information whose domain identifier is # 4 to the node # 9 (third from the top). . In this case, for example, as shown in FIG. 3, since the link j is assigned # 1 other than # 4 to the domain identifier, the node # 1 transmits the information of the link j to the node # 9. Will not. Note that filtering may be performed based on other rules.

  Next, in the network configurations of FIGS. 1A and 3, the logic by which the link DB 32 of FIG. 6B is generated by the domain identifier filtering DB 34 of FIG. 6A will be described. According to FIG. 6A, the node # 1 transmits link information whose domain identifier is # 1, # 2, # 3 or # 4 to the node # 2. That is, all link information is transmitted to the node # 2. Next, link information whose domain identifier is # 1, # 2 or # 3 is transmitted to the node # 5. That is, link information whose domain identifier is # 4 is not transmitted to the node # 5. Subsequently, link information whose domain identifier is # 4 is transmitted to the node # 9. Node # 3 also transmits link information in the same manner.

  Node # 5 transmits link information whose domain identifier is # 1, # 2 or # 3 to node # 1 and node # 4. That is, node # 5 does not transmit link information whose domain identifier is # 4. Node # 6 also transmits link information in the same manner. On the other hand, the node # 9 transmits link information whose domain identifier is # 4 to the node # 1 and the node # 8. Node # 10 also transmits link information in the same manner.

  According to the above, all link information reaches each node of the domain # 1 by exchanging link information using the link DB 32 after filtering. Next, link information belonging to the domains # 1, # 2, and # 3 arrives at the nodes of the domain # 2 and the domain # 3. Furthermore, link information belonging to domain # 4 reaches each node of domain # 4. As a result, as shown in FIG. 6B, all link information of the links a to j is stored in the link DB 32 of the nodes # 1, # 2, and # 3. The link DB 32 of the nodes # 4, # 5, # 6, and # 7 includes link information of the links a and b of the domain # 1, link information of the link c of the domain # 2, and a link of the link d of the domain # 3. Information, link information of link g between domain # 1 and domain # 2, and link information of link j between domain # 1 and domain # 3 are stored. Further, the link DB 32 of the nodes # 8, # 9, # 10, and # 11 stores the link information of the links e and f of the domain # 4.

  According to the second embodiment described above, the domain identifier filtering means 25 and the domain identifier filtering DB 34 can limit the link information that can be obtained for each domain. That is, different link information can be obtained for each domain. According to this, link information unnecessary for path setting can be removed. For example, when setting a path between nodes across domains, it is possible to pass a predetermined domain or not to pass it. become. As a result, security can be improved and maintenance and operability can be improved.

<< Third Embodiment >>
FIG. 7 is a diagram illustrating a configuration of a node according to the third embodiment. The node 2b includes a domain identifier assigning means 21, a link information exchanging means 22, a route calculating means 23, a path setting means 24, a group identifier assigning means 26, a group identifier filtering means 27, an adjacent link DB 31, a link DB 32, a path DB 33, and a group identifier. A filtering DB (group identifier filtering information storage means) 35 is included. The node 2b has a configuration in which a group identifier adding unit 26, a group identifier filtering unit 27, and a group identifier filtering DB 35 are added to the node 2 according to the first embodiment shown in FIG. Here, the group is assumed to be composed of arbitrary nodes regardless of the domain to which the group belongs.

  Similar to the domain identifier assigning means 21, the link information exchanging means 22, the route calculating means 23, and the path setting means 24, the group identifier assigning means 26 and the group identifier filtering means 27 are stored in a predetermined memory by the CPU built in the node 2b. This is realized by executing a stored program. Similar to the adjacent link DB 31, link DB 32, and path DB 33, the group identifier filtering DB 35 is constructed in a nonvolatile storage device such as a hard disk device that is built in or externally connected to the node 2b.

  The group identifier filtering DB 35 is a DB that stores information indicating which group identifier is assigned to each adjacent node. The stored information includes an adjacent node address, a domain identifier, and a target group identifier. Other information may be included.

  The adjacent node address is address information of the adjacent node. The domain identifier is an identifier of a domain to which an adjacent node belongs. The target group identifier means a group identifier of link information to be filtered and transmitted to the adjacent node.

  The group identifier assigning unit 26 assigns a group identifier to the adjacent link DB 31. Which group identifier is assigned to which link depends on the setting by the operator.

  The group identifier filtering unit 27 filters (extracts) link information in the link DB 32 using the group identifier filtering DB 35. Thereby, the adjacent node and the link information exchanging means 22 can selectively exchange link information.

  Further, as compared with FIG. 2, group identifiers are added to the adjacent link DB 31 and the link DB 32. Other means and DB are the same as those in the first embodiment.

  In the path setting process in this case, in contrast to the process shown in FIG. 4, before the link information exchanging means 22 exchanges link information with an adjacent node in step S402, the group identifier filtering means 27 obtains the link information of the link DB 32. This is a process that adds filtering.

  FIG. 8 is a diagram illustrating an example in which filtering is performed by the node 2b illustrated in FIG. 7 in the network configuration illustrated in FIG. FIG. 8A shows information of each link as a premise. FIG. 8B shows setting information of the group identifier filtering DB 35 in each node. FIG. 8C shows the link information of the link DB 32 after filtering in each node. Although FIG. 8B shows the setting information in the BN, the group identifier filtering DB 35 may be in a node other than the BN.

  Here, for a link (not shown) provided with one group identifier, filtering is performed so that link information corresponding to the target group identifier shown in FIG. 8B is transmitted from each node to an adjacent node. I do.

  In addition, for links to which a plurality of group identifiers are assigned, a link to which at least one group identifier is assigned among target group identifiers specified in the group identifier filtering DB 35 is targeted (link information is transmitted). ) Filter based on rules. Note that filtering may be performed based on other rules.

  According to FIG. 8A, for example, the link a has group identifiers # 1, # 2, and # 3. According to FIG. 8B, since the target group identifier to be transmitted to the adjacent node in each node is # 1, # 2 or # 3, the link information of link a is transmitted to all adjacent nodes in each node. Will be sent.

  The link c has group identifiers # 1 and # 2. According to the setting information in the node # 1 in FIG. 8B, for example, for the transmission from the node # 1 to the node # 2 (first from the top) or the node # 5 (second from the top), the target group Since the identifier is # 1 or # 2, it is allowed. On the other hand, transmission from node # 1 to node # 9 (third from the top) is restricted because the target group identifier is # 3.

  Next, the logic by which the link DB 32 of FIG. 8C is generated by the group identifier filtering DB 35 of FIG. 8B in the network configurations of FIG. 1A and FIG. 8A will be described. According to FIG. 8A, the links a and b have group identifiers # 1, # 2 and # 3. The links c, d, g, and h have group identifiers # 1 and # 2. Furthermore, links e, f, i and j have group identifiers # 1 and # 3. According to such group identifier assignment, all links belong to group # 1, links a, b, c, d, g and h belong to group # 2, and links a, b, e, f, i and j belongs to group # 3.

  According to FIG.8 (b), node # 1 transmits the link information whose group identifier is # 1 with respect to node # 2. Link information whose group identifier is # 2 is transmitted to the node # 5. Link information whose group identifier is # 3 is transmitted to the node # 9. Node # 3 similarly transmits link information. Next, the node # 5 transmits link information whose group identifier is # 1 to the node # 1. The link information that is the group identifier # 2 is transmitted to the node # 4. Node # 6 similarly transmits link information. Further, the node # 9 transmits link information whose group identifier is # 1 to the node # 1. The link information that is the group identifier # 3 is transmitted to the node # 8. Node # 10 similarly transmits link information.

  According to the above, all link information reaches each node of the domain # 1 by exchanging link information using the link DB 32 after filtering. Each node in domain # 2 and domain # 3 receives link information belonging to group # 2. Each node of domain # 4 receives link information belonging to group # 3. As a result, as shown in FIG. 8C, all link information of the links a to j is stored in the link DB 32 of the nodes # 1, # 2, and # 3. Further, the link information of the links a, b, c, d, g, and h of the group # 2 is stored in the link DB 32 of the nodes # 4, # 5, # 6, and # 7. Furthermore, the link DB 32 of the nodes # 8, # 9, # 10, and # 11 stores link information of a, b, e, f, i, and j of the group # 3.

  According to the third embodiment described above, the group identifier adding unit 26, the group identifier filtering unit 27, and the group identifier filtering DB 35 define a new group identifier independently of the domain identifier, By performing filtering, it is possible to limit the link information that can be obtained for each group. That is, different link information can be obtained for each group. According to this, unnecessary link information can be removed. For example, when a path is set between predetermined nodes, it is possible to pass or not pass links belonging to a predetermined group. Become. As a result, security can be improved and maintenance and operability can be improved.

<< Fourth Embodiment >>
FIG. 9 is a diagram illustrating a network configuration according to the fourth embodiment. The fourth embodiment is an application example of the third embodiment. BN (border node) has the configuration of the node 2b shown in FIG. Also, compared with FIG. 1, a physical device and a control link are added.

  A physical device (node device) can include one node or a plurality of BNs, and is physically one device. For example, nodes # 2, # 4, # 7, # 8, and # 11 are included in different physical devices, respectively. On the other hand, BN # 1, # 5, and # 9 are collectively included in one physical device. BN # 3, # 6, and # 10 are also collectively included in one physical device. It is assumed that the physical device operates as one node or BN for a link that is a data path. Accordingly, in the network shown in FIG. 9, it is possible to set a path between the domain # 1, the domain # 2, and the domain # 3.

  The control link is a communication path for exchanging link information and path setting information with an adjacent node. Information related to the control link is stored in a different DB from the link DB 32, or even if it is the same DB as the link DB 32, it is clearly distinguished that it is a control link, and is stored separately from the link information. Is done. Information about the control link is not used when calculating the path. According to this, since path setting information and link information can be exchanged between nodes via a control link different from the link that is a data path, it is independent of the data transmission / reception operation in the set path. Thus, exchange of link information and path setting can be performed, and network management and operation can be easily performed.

  FIG. 10 is a diagram showing node addresses and link information in the network configuration of FIG. In FIG. 10, “-” (hyphen) means that information is not set. For example, no domain identifier is set for the link c. This is because the link c actually belongs to the domain # 2, but the link information of the domain # 2 does not flow to other domains due to the setting of the group identifier filtering DB 35 as will be described later. Because it is not necessary to set. The same applies to the group identifier “-”.

  FIG. 11 is a diagram illustrating an example in which filtering is performed based on FIG. FIG. 11A shows setting information in each node of the assumed group identifier filtering DB 35. FIG. 11B shows the link information of the link DB 32 after filtering in each node. In FIG. 11A, “ANY” means that all link information is selected regardless of the contents of the group identifier. “NONE” means that not all link information is targeted regardless of the contents of the group identifier.

  Further, in the domain # 2 and the domain # 3, since no group identifier is used, there may be a means for deleting the group identifier in the BN.

  Next, the logic by which the link DB 32 of FIG. 11B is generated by the group identifier filtering DB 35 of FIG. 11A in the network configuration of FIGS. 9 and 10 will be described. According to FIG. 10, links a and b each have # 1 as a domain identifier and # 2 and # 3 as group identifiers. According to FIG. 11A, the node # 1 transmits all link information to the node # 2. Link information whose group identifier is # 2 is transmitted to the node # 5. Link information whose group identifier is # 3 is transmitted to the node # 9. Node # 3 also transmits link information in the same manner.

  Node # 5 does not transmit link information to node # 1. All link information is transmitted to nodes # 4 and # 6. Node # 6 also transmits link information in the same manner. Further, the node # 9 does not transmit link information to the node # 1. All link information is transmitted to the nodes # 8 and # 10. Node # 10 also transmits link information in the same manner.

  According to the above, link information belonging to the domain # 1 arrives at each node of the domain # 1 by exchanging link information using the link DB 32 after filtering. Each node in domain # 2 receives link information belonging to domain # 1 and domain # 2. Each node in domain # 3 receives link information belonging to domain # 1 and domain # 3. As a result, as shown in FIG. 11B, the link information of the links a and b of the domain # 1 is stored in the link DB 32 of the nodes # 1, # 2, and # 3. The link DB 32 of the nodes # 4, # 5, # 6 and # 7 stores the link information of the links a and b of the domain # 1, and the link information of the links c and d of the domain # 2. Further, the link DB 32 of the nodes # 8, # 9, # 10, and # 11 stores the link information of the links a and b of the domain # 1, and the link information of the links e and f of the domain # 3.

  According to the fourth embodiment described above, since the physical device for each BN can be a physical device including a plurality of BNs, the hardware cost can be reduced. Further, since the BN connecting the domains can be gathered in one place, the maintenance and operability of the network can be improved.

  The node (node device) according to the embodiment of the present invention can be realized by a computer and a program, and can be provided by recording the program on a computer-readable recording medium. Is possible. It is also possible to provide the program via a network.

<< Other embodiments >>
An example of the preferred embodiment of the present invention has been described above, but the present invention is not limited to the above-described embodiment, and can be changed as appropriate without departing from the spirit of the present invention. For example, the following embodiments can be considered.
(1) In the first embodiment, the path DB 33 has been described as a path information storage unit that stores path information such as a route. However, a memory such as a flash memory may be used as the path information storage unit. Further, an interface between programs may be used as a path information transmission unit that transmits path information from the route calculation unit 23 to the path setting unit 24.
(2) Although the fourth embodiment has been described as an application example of the third embodiment, a configuration in which a plurality of BNs are included in one physical device as illustrated in FIG. The present invention may be applied to the second embodiment. According to this, as in the fourth embodiment, the cost of hardware can be reduced, and the maintenance and operability of the network can be improved.

It is a figure which shows the structure of the network which concerns on embodiment of this invention, (a) shows the structure where several BN of a different domain exists in a physically different apparatus, (b) shows several of a different domain A configuration in which BN is physically located in the same device is shown. It is a figure which shows the structure of the node which concerns on the 1st Embodiment of this invention. It is a figure which shows the example of link DB on the assumption of the network structure of Fig.1 (a). It is a flowchart which shows the process which a node sets a path | pass. It is a figure which shows the structure of the node which concerns on the 2nd Embodiment of this invention. FIG. 6 is a diagram illustrating an example in which filtering is performed by the nodes illustrated in FIG. 5 in the network configurations of FIG. 1A and FIG. 3, and FIG. (B) shows the link information of the link DB after filtering in each node. It is a figure which shows the structure of the node which concerns on 3rd Embodiment. FIG. 8 is a diagram illustrating an example in which filtering is performed by the node illustrated in FIG. 7 in the network configuration illustrated in FIG. 1A, in which FIG. 1A illustrates information of each link that is a premise, and FIG. The setting information of group identifier filtering DB is shown, (c) shows the link information of the link DB after filtering in each node. It is a figure which shows the network structure which concerns on 4th Embodiment. FIG. 10 is a diagram showing node addresses and link information in the network configuration of FIG. 9. It is a figure which shows the example which implemented filtering on the premise of FIG. 10, (a) shows the setting information in each node of the group identifier filtering DB as a premise, (b) is the link after filtering in each node DB link information is shown.

Explanation of symbols

2, 2a, 2b nodes (node devices)
DESCRIPTION OF SYMBOLS 21 Domain identifier provision means 22 Link information exchange means 23 Path calculation means 24 Path setting means 25 Domain identifier filtering means 26 Group identifier provision means 27 Group identifier filtering means 32 Link DB (link information storage means)
34 Domain identifier filtering DB (domain identifier filtering information storage means)
35 Group identifier filtering DB (Group identifier filtering information storage means)

Claims (4)

Is operated in accordance with a predetermined policy, the nodes and the network domain there are a plurality including a link connecting between the nodes, Ri span multiple domains, the between unique address within the domain is assigned a node A node device that collects network information and sets paths.
Link information storage means for storing link information which is information of each link in the network;
Domain identifier assigning means for assigning to the link information stored in the link information storage means a domain identifier specific to the domain to which the link belongs;
And link information exchanging means for exchanging uniquely identifying link information across the network by combining the address information indicating an address within the domain of the domain identifier assigned by the domain identifier providing unit node with other nodes ,
Route calculation means for calculating a route of a path between predetermined nodes in the network using the link information;
Path setting means for transmitting and receiving path setting information including the route between nodes constituting the route according to the route calculated by the route calculating means; and
A node device comprising: Domain identifier filtering information storage means for storing domain identifier filtering information indicating a domain identifier of link information to be transmitted from the node device to each adjacent node;
Domain identifier filtering means for extracting corresponding link information from the link information storage means according to the domain identifier filtering information stored in the domain identifier filtering information storage means;
The node device according to claim 1, further comprising: Group identifier giving means for giving a group identifier unique to the group to which the link belongs to link information stored in the link information storage means;
Group identifier filtering information storage means for storing group identifier filtering information indicating a group identifier of link information to be transmitted from the node device to each adjacent node;
Group identifier filtering means for extracting corresponding link information from the link information storage means according to the group identifier filtering information stored in the group identifier filtering information storage means;
The node device according to claim 1, further comprising: A node device comprising a plurality of node devices according to any one of claims 1 to 3,
A node device comprising a control link that is a communication path for exchanging the link information and the path setting information between node devices in the node device and between node devices across the node device.

Images