Node, an optical/electrical path integrated network using the node, and a program which controls the node

1. Field of the Invention

The present invention relates to optical/electrical path integrated networks, and relates specifically to a technique for notifying about information relating to traffic quantity between sub-networks, and a technique for dynamically establishing and releasing optical paths according to the traffic quantity between sub-networks.

This application is based on patent application No. 2002-045092 and No. 2002-054247 filed in Japan, the contents of which are incorporated herein by reference.

2. Description of the Related Art

Research and development is proceeding in the field of optical/electrical path integrated networks, as a technology for building high capacity networks. In an optical/electrical path integrated network, the transmission/reception endpoints are connected by optical paths. An electrical sub-network comprising a node which performs routing based on packet header information is connected to each of the transmission/reception endpoints. A first conventional example of an optical/electrical path integrated network is shown in FIG. 6 .

The optical/electrical path integrated network comprises a photonic core network C and electrical sub-networks S 1 to S 4 . The electrical sub-networks S 1 to S 4 must be connected by optical paths for communication to occur.

The photonic core network C comprises photonic border nodes and a photonic core node. The electrical sub-networks S 1 to S 4 comprise electrical border nodes and electrical core nodes. The electrical core nodes and the photonic core node are contiguous at the boundary of the electrical sub-networks S 1 to S 4 and the photonic core network C, and are interconnected by optical fiber links.

Optical paths are established over the photonic core network C, so as to interconnect the electrical border nodes provided in the different electrical sub-networks S 1 to S 4 . Information is transferred transparently between the electrical border nodes over the optical paths.

When four optical paths are established as in FIG. 6 , the view of the electrical sub-networks S 1 to S 4 is as shown in FIG. 7 . FIG. 7 shows a view of a network topology comprising nodes which are capable of performing packet processing, in which the topology of the photonic core network C is hidden. FIG. 8 is a combined view of the electrical sub-networks S 1 to S 4 and the photonic core network C.

The electrical sub-networks must be interconnected. However, it is not necessary for all of the electrical sub-networks to be directly connected to each other by optical paths, and multiple hop routing may be used.

FIG. 9 shows an example of the connection of the four electrical sub-networks S 1 to S 4 in FIG. 6 . The electrical sub-network S 1 is connected to the electrical sub-networks S 2 and S 3 by a direct optical path, the electrical sub-network S 2 is connected to the electrical sub-networks S 1 and S 4 , the electrical sub-network S 3 is connected to the electrical sub-networks S 1 and S 4 , and the electrical sub-network S 4 is connected to the electrical sub-networks S 2 and S 3 , each by direct optical paths.

To transmit a packet from the electrical sub-network S 1 to the electrical sub-network S 4 , the packet can travel from the electrical sub-network S 1 to the electrical sub-network S 4 via the electrical sub-network S 2 , or from the electrical sub-network S 1 to the electrical sub-network S 4 via the electrical sub-network S 3 , by multi-hop routing. In the same manner as FIG. 9 , FIG. 10 shows an example of the connection of the four electrical sub-networks S 1 to S 4 in FIG. 6 . Diagonal optical paths are established between the electrical sub-networks S 1 and S 4 , and between the electrical sub-networks S 2 and S 3 .

In both FIG. 9 and FIG. 10 , the electrical border router of each electrical sub-network S 1 to S 4 has two electrical packet transmission/reception ports connected to the photonic core network C. How the two electrical packet transmission/reception ports provided in each of the electrical border routers should be directly connected by optical paths is determined by the traffic quantity between the electrical sub-networks S 1 to S 4 . FIG. 9 is favorable when the traffic quantity over the diagonal paths is small, and conversely FIG. 10 is favorable when such traffic is high.

If the optical paths are established without taking the traffic quantity into consideration, then for example electrical sub-networks which exchange a high quantity of traffic may not be directly connected by an optical path, making it necessary to transfer packets by multi-hop routing, which causes a problem of congestion of the optical paths.

Furthermore, an optical/electrical path integrated network is shown in FIG. 23 as a second conventional example. This optical/electrical path integrated network is constructed from the photonic core network C and the electrical sub-networks S 1 to S 4 . The optical/electrical path integrated network in FIG. 23 is a multi-layer network, in which optical paths are established over the photonic core network C. In this manner, the group of electrical sub-networks S 1 to S 4 , connected by optical paths, constitutes the entire electrical network.

The photonic core network C comprises photonic border nodes 1 A to 6 A and a photonic core node 7 A. The electrical sub-networks S 1 to S 4 comprise electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A and 40 A, and electrical core nodes 10 A, 20 A, 31 A, 41 A and 42 A. The electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A, 40 A and the photonic border nodes 1 A to 6 A are contiguous at the boundary between the electrical sub-networks S 1 to S 4 and the photonic core network C, and are interconnected by optical fiber links. The optical paths are established over the photonic core network C, to interconnect the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A, 40 A in the different electrical sub-networks S 1 to S 4 . Information is transmitted between the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A and 40 A transparently over the optical paths.

The topology of the electrical network can be virtually changed, depending on which of the electrical sub-networks S 1 to S 4 are connected. FIG. 24 shows how two different electrical network topologies can be realized when a single optical path network topology is applied. Furthermore, FIG. 24 describes the hierarchy of the optical paths and the electrical paths. In FIG. 24 , O-LSP (Optical-Label Switched Path) indicates an optical path, and E-LSP (Electric-Label Switched Path) indicates an electrical path.

The E-LSP is routed over the electrical network, comprising the electrical sub-networks S 1 to S 4 which are interconnected by O-LSPS. In the connection mode # 1 on the right side of FIG. 24 , the E-LSP is connected by multi-hop routing. In other words, two electrical sub-networks are connected via two O-LSPs. On the other hand, in the connection mode # 2 on the left side of FIG. 24 , the E-LSP is connected by a single hop. In other words, two electrical sub-networks are connected via a single O-LSP.

In the terminology of graph theory, the entire electrical network must be “connected”. In other words, the electrical sub-networks S 1 to S 4 must be interconnected by O-LSPs. However, it is not necessary for every one of the electrical sub-networks S 1 to S 4 to be connected to every other by an O-LSP, and multi-hop routing may also be used. FIG. 25 shows a connected electrical network and an unconnected electrical network. In the connected electrical network, all four of the electrical sub-networks S 1 to S 4 can communicate via O-LSPs, but in the unconnected electrical network only three of the electrical sub-networks are connected by O-LSPs, and one of the electrical sub-networks cannot communicate with the other three electrical sub-networks via an O-LSP.

FIG. 16 shows an example in which the four electrical sub-networks S 1 to S 4 in FIG. 25 are connected. The electrical sub-network S 1 is connected to the electrical sub-networks S 2 and S 3 by direct optical paths, the electrical sub-network S 2 is connected to the electrical sub-networks S 1 and S 4 , the electrical sub-network S 3 is connected to the electrical sub-networks S 1 and S 4 , and the electrical sub-network S 4 is connected to the electrical sub-networks S 2 and S 3 , each by direct optical paths. In order to transmit a packet from the electrical sub-network S 1 to the electrical sub-network S 4 , the packet can be transferred from the electrical sub-network S 1 to the electrical sub-network S 4 via the electrical sub-network S 2 , or from the electrical sub-network S 1 to the electrical sub-network S 4 via the electrical sub-network S 3 , by multi-hop routing.

In the same manner as FIG. 16 , FIG. 17 shows an example in which the four electrical sub-networks S 1 to S 4 in FIG. 25 are connected. Diagonal optical paths are established between the electrical sub-networks S 1 and S 4 , and between the electrical sub-networks S 2 and S 3 .

In FIG. 16 and FIG. 17 , the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A and 40 A in the electrical sub-networks S 1 to S 4 have two electrical packet transmission/reception ports connected to the photonic core network C. How the two electrical packet transmission/reception ports provided in the respective electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A and 40 A should be directly connected by optical paths is determined by the traffic quantity between the electrical sub-networks S 1 to S 4 . FIG. 16 is favorable when the traffic quantity over the diagonal paths is small, and conversely FIG. 17 is favorable when such traffic is high.

In the same manner as the first conventional example, if in this second conventional example the optical paths are established without taking the traffic quantity into consideration, then for example electrical sub-networks which exchange a high quantity of traffic may not be directly connected by an optical path, making it necessary to transfer packets by multi-hop routing, which causes a problem of congestion of the optical paths.

Traffic quantity varies temporally, and as such even once the optical paths are established, it can be necessary to dynamically reconfigure the optical paths according to the conditions. Requiring a network administrator to manually reconfigure the optical paths in this manner according to variations in traffic increases the amount of work required for maintenance, which is undesirable.

The present invention takes the above factors into consideration, with an object of providing an optical/electrical path integrated network, a node, a program and a recording medium which allows the efficient use of network resources without requiring the intervention of the network administrator, by automatically performing the establishment and release of optimal optical paths according to the traffic quantity between electrical sub-networks.

All of the electrical sub-networks are connected by optical paths so as to be connected, by either a single hop or multiple hops. Consequently, all of the electrical sub-networks are connected by electrical virtual Label Switched Paths (LSPs). The LSPs are established over the optical paths. When electrical sub-networks are connected by multiple-hop routing, the LSPs which connect these electrical sub-networks traverse a plurality of optical paths.

The electrical border node in an electrical sub-network measures the flow of packets from its own electrical sub-network which are destined for other electrical sub-networks. This measurement can be obtained by measuring the number of packets and the number of bytes transferred over the LSPs which connect these electrical sub-networks. The bit rate can be calculated by normalizing these measured values over time.

The electrical border node notifies a photonic border node of the traffic quantity measured for each LSP. Upon receiving this information, the photonic border node then notifies another photonic border node. By mutually exchanging this traffic information, the photonic border nodes can obtain information relating to the traffic quantity between the electrical sub-networks.

By thus exchanging the results of measuring traffic quantity, each photonic border node can know in an autonomous manner which electrical sub-networks are exchanging a large amount of traffic. It is possible for the photonic border nodes to establish or release optical paths in an autonomous manner, based on these pieces of information.

If information relating to the optical paths which have been established or released is then exchanged between the photonic border nodes, and the photonic border nodes notify the electrical border nodes of this information, the electrical border nodes can become aware that the topology of the electrical sub-networks has changed, and change the paths of the LSPs accordingly. FIG. 4 conceptually shows the dynamic establishment or release of optical paths according to traffic quantity.

In other words, a first aspect of the present invention is an optical/electrical path integrated network, which comprises a plurality of electrical sub-networks comprising nodes which perform routing based on packet header information, and a photonic core network comprising nodes which are interconnected by optical paths, wherein electrical border nodes and photonic border nodes are interconnected between the photonic core network and the plurality of electrical sub-networks, respectively, by optical paths.

Here, a characteristic of the present invention is that the electrical border nodes comprise a measuring device which measures the traffic quantity between its own electrical sub-network and other electrical sub-networks, and a device which notifies a photonic border node about the information relating to traffic quantity obtained by the measurement results from the measuring device, and that the photonic border nodes comprise a exchanging device which exchanges with other photonic border nodes the information relating to traffic quantity received from the electrical border nodes, a device which gathers information relating to traffic quantity between the plurality of electrical sub-networks based on the plurality of pieces of information relating to traffic quantity which are exchanged by the exchanging device, and a gathering device which sets up the connection mode between the electrical sub-networks, based on the information relating to traffic quantity gathered by the gathering device.

The device which sets up the connection mode preferably comprises a device which sets the distance between electrical sub-networks, between which traffic is generated, in inverse proportion to the quantity of traffic.

In addition, the photonic border nodes preferably comprise an exchanging device which when the connection mode between the electrical sub-networks has been changed by the device which sets up the connection mode, exchanges optical path establishment change information containing information detailing this change with other photonic border nodes, a gathering device which gathers network topology information based on a plurality of pieces of the optical path establishment change information exchanged by this exchanging device, and a device which notifies the electrical border nodes about the network topology information gathered by this gathering device.

A second aspect of the present invention is an electrical border node for use with the optical/electrical path integrated network, which comprises a plurality of electrical sub-networks, comprising nodes which perform routing based on packet header information, and a photonic core network comprising nodes which are interconnected by optical paths, wherein the electrical border nodes and the photonic border nodes are interconnected between the photonic core network and the plurality of electrical sub-networks, respectively, by optical paths.

Here, a characteristic of the present invention is that it comprises a device which measures its own traffic quantity between the electrical sub-network and other electrical sub-networks, and a device which notifies the photonic border nodes about information relating to traffic quantity, which is based on the results of the measuring device.

A third aspect of the present invention is an electrical border node for use with the optical/electrical path integrated network which comprises a plurality of electrical sub-networks comprising nodes which perform routing based on packet header information, and a photonic core network comprising nodes which are interconnected by optical paths, wherein the electrical border nodes and the photonic border nodes are interconnected between the photonic core network and the plurality of electrical sub-networks, respectively, by optical paths, and the electrical border node comprises a device which measures the traffic quantity between its own electrical sub-network and other electrical sub-networks, and a device which notifies the photonic border nodes about the information relating to traffic quantity which is based on the results obtained by the measuring device.

Here, a characteristic of the present invention is that it comprises an exchanging device which exchanges the information relating to traffic quantity received from the electrical border nodes with other photonic border nodes, a gathering device which gathers information relating to the traffic quantity between the plurality of electrical sub-networks based on a plurality of the pieces of information relating to traffic quantity exchanged by this exchanging device, and a device which sets up the connection mode between the electrical sub-networks based on the information relating to traffic quantity gathered by the gathering device.

The device which sets up the connection mode preferably comprises a device for setting the distance between electrical sub-networks between which traffic is generated, in inverse proportion to the quantity of traffic.

In addition, it preferably comprises an exchanging device which when the connection mode between the electrical sub-networks has been changed by the device which sets up the connection mode, exchanges optical path establishment change information containing information detailing this change with other photonic border nodes, a device which gathers network topology information based on a plurality of pieces of the optical path establishment change information exchanged by this exchanging device, and a device which notifies the electrical border nodes about the network topology information gathered by the gathering device.

A fourth aspect of the present invention is a program, which by its installation on an information processing apparatus enables the information processing apparatus to realize the functions required of an apparatus which controls the electrical border node for use with the optical/electrical path integrated network which comprises a plurality of electrical sub-networks comprising nodes which perform routing based on packet header information and a photonic core network comprising nodes which are interconnected by optical paths, wherein the electrical border nodes and the photonic border nodes are interconnected between the photonic core network and the plurality of electrical sub-networks, respectively, by optical paths.

Here, the program of the present invention realizes a measuring function which measures the traffic quantity between its own electrical sub-network and other electrical sub-networks, and a function which notifies the photonic border nodes about this information relating to traffic quantity which is based on the measurement results of this measuring function.

Alternatively, the fourth aspect of the present invention is a program which by its installation on an information processing apparatus enables the information processing apparatus to realize the functions required of an apparatus which controls an electrical border node for use with the optical/electrical path integrated network which comprises a plurality of electrical sub-networks comprising nodes which perform routing based on packet header information and a photonic core network comprising nodes which are interconnected by optical paths, in which the electrical border nodes and the photonic border nodes are interconnected between the photonic core network and the plurality of electrical sub-networks, respectively, by optical paths, wherein the electrical border nodes comprise a function which measures the traffic quantity between its own electrical sub-network and other electrical sub-networks, and a function which notifies the photonic border nodes about the information relating to traffic quantity which is based on the measurement results of the measuring function.

Here, the program of the present invention realizes an exchanging function which exchanges the information relating to traffic received from the electrical border nodes with other photonic border nodes, a gathering function which gathers information relating to traffic quantity between the plurality of electrical sub-networks based on the plurality of pieces of information relating to traffic quantity which are exchanged by this exchanging function, and a function which sets up the connection mode between the electrical sub-networks based on the information relating to traffic gathered by this gathering function.

The function which sets up the connection mode preferably realizes a function which sets the distance between electrical sub-networks between which traffic is generated, in inverse proportion to the traffic quantity.

In addition, the program of the present invention preferably realizes an exchanging function which when the connection mode between the electrical sub-networks has been changed by the function which sets up the connection mode, exchanges optical path establishment change information containing information detailing the change with other photonic border nodes, a gathering function which gathers network topology information based on a plurality of pieces of the optical path establishment and release information exchanged by this exchanging function, and a function which notifies the electrical border nodes about the network topology information gathered by this gathering device.

A fifth aspect of the present invention is a storage medium on which the program of the present invention is stored, which is capable of being read by the aforementioned information processing device. By storing the program of the present invention on the storage medium of the present invention, the information processing apparatus can install the program of the present invention using this storage medium. Alternatively, the program of the present invention may be installed on the information processing apparatus over a network, from a server on which the program of the present invention is stored.

Accordingly, by automatically performing the establishment or release of optimal optical paths according the traffic quantity between the electrical sub-networks using an information processing apparatus such as a computer, an optical/electrical path integrated network and a node which can make effective use of network resources without requiring the intervention of the network administrator can be realized.

A main characteristic of the present invention is in an algorithm used to automatically establish or release optimal optical paths according to the traffic quantity between electrical sub-networks.

First, in a connection phase, the electrical sub-networks exchanging high actual quantity of traffic are directly connected by optical paths. Then in a capacity verification phase, virtual routing is performed, and electrical sub-networks between which congestion is likely to occur are sought and directly connected by optical paths. In addition, according to the actual optical path usage, for an optical path with the usage ratio equal to or greater than a threshold value α, additional optical paths are established so that the electrical paths which go through this optical path can travel between the electrical sub-networks via a single hop. Furthermore, according to the actual optical path usage, for an optical path with the usage ratio equal to or below a threshold value β, the electrical paths established on this optical paths are virtually bypassed to other optical paths, and if as a result there is determined to be no possibility of congestion occurring, the optical path is released.

By automatically establishing or releasing optimal optical paths according to the traffic quantity between electrical sub-networks using such an algorithm, it is possible to make effective use of network resources without requiring the intervention of the network administrator.

In other words, a sixth aspect of the present invention is an optical/electrical path integrated network comprising electrical sub-networks which exchange data in packet units and a photonic core network which interconnects these electrical sub-networks, wherein the photonic core network comprises photonic border nodes and a photonic core node, and the electrical sub-networks comprise electrical border nodes and electrical core nodes, and the electrical border nodes and the photonic border nodes provided in the electrical sub-networks and the photonic core network which are contiguous are directly connected.

Here, a characteristic of the present invention is that the photonic border nodes comprise a storage device which stores topology information for the photonic core network, a calculating device which calculates the shortest path between photonic border nodes based on the topology information stored in the storage device, and a device which establishes optical paths over this shortest distance calculated by the calculating device, and that the electrical border nodes comprise a storage device which stores the topology information for the network which is constructed from the optical paths established over the photonic core network, a calculating device which calculates the shortest path between electrical border nodes based on the topology information stored in the storage device, and a device which establishes electrical paths over this shortest path calculated by this calculating device, and that a detecting device which detects the two electrical sub-networks which exchange the highest quantity of traffic and are not yet directly connected by an optical path is provided, and the device which establishes the optical paths comprises an optical path establishment device which establishes optical paths between the two electrical sub-networks detected by the detecting device.

In addition, the device which establishes the electrical paths preferably comprises an electrical path establishment device which establishes electrical paths over the shortest path between the electrical sub-networks on the optical paths established by the optical path establishment device, and further comprises a device which performs virtual routing over the electrical paths established by this electrical path establishment device, a congested optical path detection device which detects congested locations on the optical paths based on the results of the virtual routing by the device which performs virtual routing, and a new optical path establishing device, which when a congested optical path as detected by this congested optical path detection device does not directly connect the electrical sub-networks which are causing the congestion, newly establishes an optical path which directly connects these electrical sub-networks.

Furthermore, it preferably comprises a determining device which determines whether or not the usage ratio of the optical paths established by either the optical path establishment device or the new optical path establishment device is equal to or greater than a threshold value α, and further comprises a multi-hop electrical path detection device which detects those electrical paths which do not directly connect electrical sub-networks which travel optical paths which have the usage ratio equal to or greater than the threshold value α based on the results of the determining device, a selection device which selects the electrical path which carries the highest traffic quantity from the electrical paths detected by this multi-hop electrical path detection device, and an optical path adding device which establishes an optical path so that the electrical path selected by this selection device directly connects the electrical sub-networks.

Furthermore, it is also preferable to comprise a determining device which determines whether or not the usage ratio of the optical paths established by either the optical path establishment device or the new optical path establishment device or the optical path adding device is equal to or below a threshold value β, and to further comprise a bypassing device which virtually bypasses the electrical paths established over the optical paths which are determined by this determining device to have the usage ratio equal to or below the threshold value β to other optical paths, a congested optical path detection device which detects congested locations on the optical paths after this virtual bypassing is performed by the bypassing device, and a device which actually releases the optical paths which have the usage ratio equal to or below the threshold value β after the virtual bypassing has actually been performed by the bypassing device when congested optical paths are not detected by the congested optical path detection device.

A seventh aspect of the present invention is a node for use in the optical/electrical path integrated network of the present invention, comprising at least one of a storage device which stores topology information for the photonic core network, a calculating device which calculates the shortest path between photonic border nodes based on the topology information stored in the storage device, a device which establishes an optical path over this shortest path calculated by the calculating device, a storage device which stores the topology information for the network constructed from the optical paths established over the photonic core network, a calculating device which calculates a shortest path between electrical border nodes based on the topology information stored in the storage device, a device which establishes electrical paths over this shortest path calculated by this calculating device, and a detecting device which detects the two electrical sub-networks which exchange the highest quantity of traffic and are not yet directly connected by an optical path, wherein the device which establishes the optical paths comprises an optical path establishment device which establishes optical paths between the two electrical sub-networks detected by the detecting device.

In addition, the device which establishes the electrical paths preferably comprises an electrical path establishment device which establishes electrical paths over the shortest path between the electrical sub-networks on the optical paths established by the optical path establishment device, and further comprises a device which performs virtual routing over the electrical paths established by this electrical path establishment device, a congested optical path detection device which detects congested locations on the optical paths based on the results of the device which performs virtual routing, and a new optical path establishment device, which when a congested optical path as detected by this congested optical path detection device does not directly connect the electrical sub-networks which are causing the congestion, newly establishes an optical path which directly connects these electrical sub-networks.

Furthermore, it preferably comprises a determining device which determines whether or not the usage ratio of the optical paths established by either the optical path establishment device or the new optical path establishment device is equal to or greater than a threshold value α, and further comprises a multi-hop electrical path detection device which detects those electrical paths which do not directly connect between electrical sub-networks which travel optical paths which have the usage ratio equal to or greater than the threshold value α based on the results of the determining device, a selection device which selects the electrical path which carries the highest traffic quantity from the electrical paths detected by this multi-hop electrical path detection device, and an optical path adding device which establishes an optical path so that the electrical path selected by this selection device directly connects the electrical sub-networks.

Furthermore, it is also preferable to comprise a determining device which determines whether or not the usage ratio of the optical paths established by either the optical path establishment device or the new optical path establishment device is equal to or below a threshold value β, and further comprises a bypassing device which virtually bypasses the electrical paths established over the optical paths which are determined by this determining device to have the usage ratio equal to or below the threshold value β to other optical paths, a congested optical path detection device which detects congested locations on the optical paths after this virtual bypassing is performed by the bypassing device, and a device which actually releases the optical paths which have the usage ratio equal to or below the threshold value β after the virtual bypassing has actually been performed by the bypassing device when congested optical paths are not detected by the congested optical path detection device.

An eighth aspect of the present invention is a program which by its installation on an information processing apparatus enables the information processing apparatus to realize the functions required of an apparatus which controls the node for use in the optical/electrical path integrated network of the present invention, characterized in realizing at least one of a storage function which stores the topology information for the photonic core network, a calculating function which calculates the shortest path between photonic border nodes based on the topology information stored by the storage function, a calculating function which establishes an optical path over this shortest path calculated by the calculating function, a storage function which stores the topology information for the network constructed from the optical paths established over the photonic core network, a function which calculates the shortest path between electrical border nodes based on the topology information stored by the storage function, a function which establishes an electrical path over this shortest path calculated by this calculating function, and a detecting function which detects the two electrical sub-networks which exchange the highest quantity of traffic between the electrical sub-networks and are not yet directly connected by an optical path, wherein an optical path establishment function establishes optical paths between the two electrical sub-networks detected by the detecting function is realized.

In addition, the function which establishes the electrical paths preferably realizes an electrical path establishment function which establishes an electrical path over the shortest path between the electrical sub-networks on the optical paths established by the optical path establishment function, and further realizes a function which performs virtual routing over the electrical path established by this electrical path establishment function, a congested optical path detection function which detects the congested locations on the optical paths based on the results of the virtual routing function, and a new optical path establishment function, which when a congested optical path as detected by this congested optical path detection function does not directly connect the electrical sub-networks which are causing the congestion, newly establishes an optical path which directly connects these electrical sub-networks.

Furthermore, it preferably realizes a determining function which determines whether or not the usage ratio of the optical paths established by either the optical path establishment function or the new optical path establishment function is equal to or greater than a threshold value α, and further realizes a multi-hop electrical path detection function which detects those electrical paths which do not directly connect between electrical sub-networks which travel optical paths which have the usage ratio equal to or greater than the threshold value α based on the results of the determining device, a function which selects the electrical path which carries the highest traffic quantity from the electrical paths detected by this multi-hop electrical path detection function, and an optical path adding function which establishes an optical path so that the electrical path selected by this selection function directly connects the electrical sub-networks.

Furthermore, it is also preferable to realize a determining function which determines whether or not the usage ratio of the optical paths established by either the optical path establishment function or the new optical path establishment function or the optical path adding function is equal to or below a threshold value β, and to further realize a bypassing function which virtually bypasses the electrical paths established over the optical paths which are determined by this determining function to have the usage ratio equal to or below the threshold value β to other optical paths based on the results from the determining function, a congested optical path detection function which detects congested locations on the optical paths after this virtual bypassing is performed by the bypassing function, and a function which actually releases the optical paths which have the usage ratio equal to or below the threshold value β after the virtual bypassing has actually been performed by the bypassing function when congested optical paths are not detected by the congested optical path detection function.

A ninth aspect of the present invention is a storage medium on which the program of the present invention is stored, which is capable of being read by the aforementioned information processing device. By storing the program of the present invention on the storage medium of the present invention, the information processing apparatus can install the program of the present invention using this storage medium. Alternatively, the program of the present invention may be installed on the information processing apparatus over a network, from a server on which the program of the present invention is stored.

Accordingly, by automatically performing the establishment or release of optimal optical paths according the traffic quantity between the electrical sub-networks using an information processing apparatus such as a computer, an optical/electrical path integrated network and a node which can make effective use of network resources without requiring the intervention of the network administrator can be realized.

A tenth aspect of the present invention is a path establishment method for use with the optical/electrical path integrated network of the present invention, characterized in that the photonic border nodes store the topology information for the photonic core network, calculate the shortest path between the photonic border nodes based on this stored topology information, and establish optical paths over this calculated shortest path, and the electrical border nodes store the topology information for the network constructed from the optical paths established over the photonic core network, calculate the shortest path between electrical border nodes based on the stored topology information, establish electrical paths over this calculated shortest path, detect the two electrical sub-networks which exchange the highest quantity of traffic and are not yet directly connected by an optical path, and establish optical paths between the two electrical sub-networks detected.

In addition, it is preferable that electrical paths are established over the shortest paths between the electrical sub-networks on the established optical paths, virtual routing is performed over these established electrical paths, congested locations on the optical paths are detected based on the results of this virtual routing, and that when a congested optical path as shown by the detection results does not directly connect the electrical sub-networks which are causing the congestion, an optical path which directly connects these electrical sub-networks is newly established.

Furthermore, it is preferable that a determination is made as to whether the usage ratio of the established optical paths is equal to or above a threshold value α, that electrical paths which do not directly connect between electrical sub-networks and which travel optical paths which have the usage ratio equal to or greater than the threshold value α are detected based on the determination results, that the electrical path with the highest traffic quantity is selected from these detected electrical paths, and that an optical path is then established so that this selected electrical path directly connects the electrical sub-networks.

Furthermore, it is preferable that a determination is made as to whether the usage ratio of the established optical paths is equal to or below a threshold value β, that electrical paths established on optical paths which are shown to have the usage ratio equal to or below the threshold value β are virtually bypassed to other optical paths based on the determination results, that congested locations upon the optical paths are detected after this virtual bypassing has occurred, and that when congested optical paths are not detected based on these detection results, the optical paths which after actually performing this virtual bypassing have the usage ratio equal to or below the threshold value β are actually released.

As described above, according to the present invention, by automatically establishing or releasing optimal optical paths according to the traffic quantity between electrical sub-networks, it is possible to make effective use of network resources, without requiring the intervention of the network administrator.

FIG. 1 is a diagram showing an optical/electrical path integrated network of an embodiment of the present invention.

FIG. 2 is a block diagram showing an electrical border node of the present embodiment.

FIG. 3 is a block diagram showing a photonic border node of the present embodiment.

FIG. 4 is a diagram showing a connection mode # 1 and a connection mode # 2 of the present embodiment.

FIG. 5 is a diagram describing a BGP-4 protocol used in the present embodiment.

FIG. 6 is a diagram showing the relationship between the photonic core network and the electrical sub-networks.

FIG. 7 is a view showing only the electrical sub-networks.

FIG. 8 is a combined view of the electrical sub-networks and the photonic core network.

FIG. 9 is a diagram showing an example of a connection mode of the electrical sub-networks.

FIG. 10 is a diagram showing an example of a connection mode of the electrical sub-networks.

FIG. 11 is a diagram outlining the propagation of traffic information in the present embodiment.

FIG. 12 is a block diagram showing a photonic border node of the present embodiment.

FIG. 13 is a block diagram showing an electrical border node of the present embodiment.

FIG. 14 is a flowchart showing the operational steps in a connection phase of the present embodiment.

FIG. 15 is a flowchart showing the operational steps in a capacity verification phase of the present embodiment.

FIG. 16 is a diagram showing a connection mode # 1 of the electrical sub-networks.

FIG. 17 is a diagram showing a connection mode # 2 of the electrical sub-networks.

FIG. 18 is a diagram showing a traffic matrix.

FIG. 19 is a diagram showing an E-LSP hop count matrix.

FIG. 20 is a diagram showing an example of a shortest path of an O-LSP.

FIG. 21 is a flowchart showing an additional O-LSP establishing procedure of the present embodiment.

FIG. 22 is a flowchart showing an O-LSP releasing procedure of the present embodiment.

FIG. 23 is a diagram showing the entire structure of the optical/electrical path integrated network.

FIG. 24 is a diagram showing a connection mode of the electrical network.

FIG. 25 is a diagram describing a connected electrical network and an unconnected electrical network.

As follows is a description of an optical/electrical path integrated network of an embodiment of the present invention, referring to FIG. 1 through FIG. 5 . FIG. 1 is a diagram showing an optical/electrical path integrated network of a first embodiment of the present invention. FIG. 2 is a block diagram showing an electrical border node of the first embodiment of the present invention. FIG. 3 is a block diagram showing a photonic border node of the first embodiment. FIG. 4 is a diagram showing a connection mode # 1 and a connection mode # 2 of the first embodiment. FIG. 5 is a diagram describing the BGP-4 protocol used in the first embodiment.

As shown in FIG. 1 , the first embodiment of the present invention is an optical/electrical path integrated network comprising a plurality of electrical sub-networks S 1 to S 4 which comprise nodes which perform routing based on packet header information and a photonic core network C which comprises nodes which are interconnected by optical paths, wherein the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 and the photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 are interconnected between the photonic core network C and the electrical sub-networks S 1 to S 4 by optical paths.

Here, a characteristic of the present invention is that as shown in FIG. 2 , the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 comprise a traffic measuring section 50 which measures the traffic quantity between its own electrical sub-network to which each electrical border node belongs and other electrical sub-networks, and a traffic information compiling section 51 which notifies the photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 about the information relating to traffic quantity based on the measurement results of the traffic measuring section 50 , and that as shown in FIG. 3 the photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 comprise a traffic information exchanging section 60 which exchanges the information relating to traffic quantity received from the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 with other photonic border nodes, a traffic information gathering section 61 which gathers information relating to the traffic quantity between the electrical sub-networks S 1 to S 4 based on the plurality of pieces of information relating to traffic quantity which are exchanged by the traffic information exchanging section 60 , and an optical path establishing section 62 which establishes the connection mode between the electrical sub-networks S 1 to S 4 based on the information relating to traffic quantity gathered by the traffic information gathering section 61 .

As shown in FIG. 4 , the optical path establishing section 62 establishes the distance between electrical sub-networks between which traffic is generated, in inverse proportion to the quantity of traffic. In the example shown in FIG. 4 , when the traffic between the electrical sub-networks S 1 to S 4 is approximately even, the connection mode # 1 is set up, and when there is a large quantity of traffic between the electrical sub-networks S 1 and S 4 , and between the electrical sub-networks S 2 and S 3 , the connection mode # 2 is set up.

Therefore, when there is a large quantity of traffic between the electrical sub-networks S 1 and S 4 , and between the electrical sub-networks S 2 and S 3 , direct connections are established between the electrical sub-networks S 1 to S 4 , thereby preventing congestion from occurring.

In addition, the photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 comprise a network topology information gathering section 63 which when the connection mode between the electrical sub-networks S 1 to S 4 has been changed by the optical path establishing section 62 , exchanges optical path establishment change information detailing the changes with other photonic border nodes, and gathers network topology information based on the plurality of pieces of optical path establishment change information which have been exchanged, and notifies the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 about this gathered network topology information.

As a result, the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 are made aware that the topology of the electrical sub-networks S 1 to S 4 has changed, and can change the path of an LSP accordingly. The dynamic establishment and release of optical paths based on traffic is outlined in FIG. 4 .

A control device for the optical/electrical path integrated network of the present embodiment, or a control device for the electrical border nodes, a control device for the electrical core nodes, a control device for the photonic border nodes, and a control device for the photonic core node, can be realized using a computer, which is an information processing device. In other words, the program of the present embodiment by its installation onto a computer can allow the computer to perform the functions required of an apparatus which controls the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 of the present embodiment, and realizes a function corresponding with the traffic measuring section 50 which measures the traffic quantity between the electrical sub-network to which each electrical border node belongs and other electrical sub-networks, and a function corresponding with the traffic information compiling section 51 which notifies the photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 about the information relating to traffic which is based on the measurement results of the traffic measuring section 50 .

In addition, the program of the present embodiment by its installation onto a computer can allow the computer to perform the functions required of an apparatus which controls the photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 of the present embodiment, and realizes a function corresponding with the traffic information exchanging section 60 which exchanges the information relating to traffic quantity received from the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 with other photonic border nodes, a function corresponding with the traffic information gathering section 61 which gathers information relating to the traffic between electrical sub-networks S 1 to S 4 based on the plurality of pieces of information relating to traffic quantity which are exchanged by the traffic information exchanging section 60 , and a function corresponding with the optical path establishing section 62 which establishes the connection mode between the electrical sub-networks S 1 to S 4 based on the information relating to traffic quantity gathered by the traffic information gathering section 61 , and furthermore allows the function corresponding with this optical path establishing section 62 to establish the distance between electrical sub-networks, between which traffic is generated, in inverse proportion to the quantity of traffic; and further realizes a function corresponding with the network topology information gathering section 63 which when the connection mode between the electrical sub-networks S 1 to S 4 has been changed by the optical path establishing section 62 exchanges optical path establishment change information detailing the changes with other photonic border nodes, and gathers network topology information based on the plurality of pieces of optical path establishment change information which have been exchanged, and notifies the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 about this gathered network topology information.

In addition, the control devices for the electrical core nodes and the photonic core node can also be realized using a computer.

Furthermore, by storing the program of the present embodiment on the storage medium of the present invention, the computer can install the program of the present embodiment using this storage medium. Alternatively, the program of the present embodiment may be installed directly on the information processing apparatus over a network, from a server on which the program of the present embodiment is stored.

Accordingly, by automatically performing the establishment or release of optimal optical paths according the traffic quantity between the electrical sub-networks using a computer, an optical/electrical path integrated network and a node which can make effective use of network resources without requiring the intervention of the network administrator can be realized.

The first embodiment is described in more detail below.

An example is used in which the exchange of traffic quantity information between photonic border nodes, and between photonic border nodes and electrical border nodes, is performed using a modified version of a standard Internet protocol.

The BGP-4 protocol is a standard Internet protocol. This protocol is used to exchange path information between autonomous systems (AS). First a session is established between border routers in an autonomous system, and then path information is exchanged. A session established between border routers in different autonomous systems is called an E-BGP (External BGP) session, and a session established between border routers belonging to the same AS is called an I-BGP (Internal BGP) session.

FIG. 5 outlines an I-BGP session and an E-BGP session. In the I-BGP session, the border routers within the same AS are connected in a full mesh configuration. Information about reachable paths is transmitted between autonomous systems using an E-BGP session. The path information transmitted in the E-BGP session is then transmitted to every border router within the same AS using an I-BGP session.

Using this transmission scheme, information on traffic quantity is exchanged between photonic border routers, and between photonic border nodes and electrical border nodes. FIG. 1 outlines the propagation of traffic information in the present embodiment. The traffic information measured at the electrical border nodes 11 , 12 , 21 , 22 , 30 , 32 , 40 is transmitted to the photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 via an E-BGP session. The photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 which receive this traffic information then transmit the traffic information to the other photonic border nodes within the photonic core network C via an I-BGP session. Each of the photonic border nodes 1 , 2 , 3 , 4 , 5 , 6 then gathers traffic information for the LSPs between the electrical sub-networks S 1 to S 4 based on the received traffic quantity information, and in order to set the distance between electrical sub-networks between which traffic is generated, in inverse proportion to the quantity of traffic, performs the establishment or release of optical paths in an autonomous manner.

Once the photonic border nodes have performed the establishment or release of the optical paths, the information about the optical paths (which electrical sub-networks they connect between) is transmitted using BGP sessions. This information is transmitted to the electrical border nodes via an I-BGP session, and to the photonic border nodes via an E-BGP session.

Furthermore, an example is described in which the exchange of traffic quantity information between photonic border nodes, and between photonic border nodes and electrical border nodes, is performed using a modified version of a standard Internet protocol called OSPF (Open Shortest Path First).

OSPF is a standard Internet protocol. This protocol allows the nodes within an autonomous system to exchange link-states. By having a node which generates link information advertise a packet known as a link-state packet to an adjacent node, and having this node then advertise a link state packet to an adjacent node, every node within the autonomous system can become aware of the link-states. In this manner, it is possible to share link-states between every node within the network using OSPF.

Using this transmission scheme, information on traffic quantity is exchanged between photonic border routers, and between photonic border nodes and electrical border nodes. FIG. 11 outlines the transmission of traffic information in the present embodiment.

An electrical border node 140 appends traffic information to a link-state packet and advertises this link-state packet to a photonic border node 103 . Upon receipt of this packet, the photonic border node 103 then advertises the link-state packet to the adjoining photonic nodes 102 , 107 , 106 . These photonic nodes then advertise the link-state packet to any adjacent photonic nodes. When the photonic nodes receive further copies of a link-state packet which has already been received, the photonic nodes do not perform any further advertising of the link-state packet. By repeating this process, it is possible for each photonic border node and electrical border node to gather information on the traffic between each of the electrical sub-networks. Furthermore, by using a modified version of OSPF, it is possible for each photonic border node and electrical border node to gather information on the traffic between each of the electrical sub-networks.

A second embodiment of an optical/electrical path integrated network is described below, referring to FIG. 12 through FIG. 22 . FIG. 12 is a block diagram showing a photonic border node of the second embodiment. FIG. 13 is a block diagram showing an electrical border node of the second embodiment. FIG. 14 is a flow chart showing the operational steps in a connection phase of the present embodiment. FIG. 15 is a flowchart showing the operational steps in a capacity verification phase of the present embodiment. FIG. 16 is a diagram showing a connection mode # 1 of the electrical sub-networks. FIG. 17 is a diagram showing a connection mode # 2 of the electrical sub-networks. FIG. 18 is a diagram showing a traffic matrix. FIG. 19 is a diagram showing an E-LSP hop count matrix. FIG. 20 is a diagram showing an example of a shortest path of an O-LSP. FIG. 21 is a flowchart showing an additional O-LSP establishing procedure of the present embodiment. FIG. 22 is a flowchart showing an O-LSP releasing procedure of the present embodiment. Furthermore, the entire construction of the optical/electrical path integrated network is shown in FIG. 23 .

The optical/electrical path integrated network according to the second embodiment comprises electrical sub-networks S 1 to S 4 which exchange data in packet units, and a photonic core network C which interconnects these electrical sub-networks S 1 to S 4 , in which the photonic core network C comprises photonic border nodes 1 to 6 and a photonic core node 7 , and the electrical sub-networks S 1 to S 4 comprise electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A, 40 A, and electrical core nodes 10 , 20 , 31 , 41 , 42 , and the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A and 40 A and the photonic border nodes 1 A to 6 A provided in the electrical sub-networks S 1 to S 4 and the photonic core network C which are contiguous are directly connected.

Here, a characteristic of the second embodiment is that the photonic border nodes 1 A to 6 A comprise a topology information storage section 50 A which stores the topology information of the photonic core network C, an optical path shortest path calculation section 51 A which calculates a shortest path between the photonic border nodes 1 A to 6 A based on the topology information stored in the topology information storage section 50 A, and an optical path establishment/release section 52 A which establishes an optical path over this shortest path calculated by the optical path shortest path calculation section 51 A, and that the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A, 40 A comprise a topology information storage section 60 A which stores the topology information of the network constructed from the optical paths established over the photonic core network C, an electrical path shortest path calculation section 61 A which calculates the shortest path between the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A and 40 A based on the topology information stored in the topology information storage section 60 A, and an electrical path establishing section 62 A which establishes an electrical path over the shortest path calculated by the electrical path shortest path calculation section 61 A, and is further characterized in that a traffic quantity information measuring section 64 A which detects the two electrical sub-networks of the electrical sub-networks S 1 to S 4 which exchange the highest quantity of traffic and are not yet connected by an optical path, a traffic quantity information notification section 63 A, and a traffic information gathering section 53 A are provided, and in that the optical path establishment/release section 52 A establishes an optical path between the two electrical sub-networks detected by the traffic information gathering section 53 A.

In addition, the electrical path establishing sections 62 A in the electrical border nodes comprise a virtual routing execution section 65 A which establishes electrical paths over the shortest paths between the electrical sub-networks S 1 to S 4 on the optical paths established by the optical path establishment/release section 52 A, and performs virtual routing via these established electrical paths, and the photonic border nodes comprise congested optical path detection sections 54 A which detect the congested locations on the optical paths after this virtual bypassing is performed based on the results of the virtual routing performed by the virtual routing execution section 65 A, and the optical path establishment/release section 52 A, which when a congested optical path as detected by this congested optical path detection section 54 A does not directly connect the electrical sub-networks which are causing the congestion, newly establishes an optical path which directly connects these electrical sub-networks.

Furthermore, the photonic border node preferably comprises an optical path usage ratio determining section 55 A which determines whether or not the usage ratio of the optical paths is equal to or greater than a threshold value α, and further comprises a multi-hop electrical path detection section 56 A which detects those electrical paths which do not directly connect between electrical sub-networks which travel optical paths which have the usage ratio equal to or greater than the threshold value α based on the results of the optical path usage ratio determining section 55 A, and a highest traffic electrical path selection section 57 A which selects the electrical path which carries the highest traffic quantity from the electrical paths detected by this multi-hop electrical path detection section 56 A, and the optical path establishment/release section 52 A establishes an optical path so that the electrical path selected by the highest traffic electrical path selection section 57 A directly connects the electrical sub-networks S 1 to S 4 .

Furthermore, the optical path usage ratio determining section 55 A determines whether or not the usage ratio of the established optical paths is equal to or below a threshold value β, and the photonic border node comprises a virtual bypass processing section 58 A which virtually bypasses electrical paths established over optical paths which have the usage ratio equal to or below the threshold value β to other optical paths based on this determination, and the congested optical path detection section 54 A detects congested locations upon the optical paths after the virtual bypassing is performed by the virtual bypass processing section 58 A, and the optical path establishment/release section 52 A actually releases the optical paths which have the usage ratio equal to or below the threshold value β after the virtual bypassing has actually been performed, when congested optical paths are not detected by the congested optical path detection section 54 A.

The apparatus which controls the photonic border nodes 1 A to 6 A and the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A and 40 A is realized using a computer. In other words, by installing a program on a computer which enables the computer to realize the functions required of an apparatus which controls the nodes for use with the optical/electrical path integrated network of the present embodiment, it is possible to obtain a device which controls the photonic border nodes 1 A to 6 A and the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A, 40 A, wherein the program realizes at least one of; a function corresponding with the topology information storage section 50 A which stores the topology information of the photonic core network C, a function corresponding with the optical path shortest path calculation section 51 A which calculates the shortest path between the photonic border nodes 1 A to 6 A based on the topology information stored in the topology information storage section 50 A, a function corresponding with the optical path establishment/release section 52 A which establishes an optical path over the shortest path calculated by the optical path shortest path calculation section 51 A, a function corresponding with the topology information storage section 60 A which stores the topology information of the network constructed from the optical paths established on the photonic core network C, a function corresponding with the electrical path shortest path calculation section 61 A which calculates the shortest path between the electrical border nodes based on the topology information stored in the topology information storage section 60 A, a function corresponding with the electrical path establishing section 62 A which establishes an electrical path over the shortest path calculated by the electrical path shortest path calculation section 61 A, and functions corresponding with the traffic measuring section 64 A which detects the two electrical sub-networks not yet directly connected by electrical paths which exchange the highest quantity of traffic, the traffic quantity information notification section 63 A, and the traffic information gathering section 53 A. Moreover this also realizes an optical path establishment function which establishes an optical path between the two electrical sub-networks detected by the traffic information gathering section 53 A, as a function corresponding with the optical path establishment/release section 52 A. Furthermore this realizes an electrical path establishment function which establishes electrical paths over the shortest paths between the electrical sub-networks on the optical path established by the optical path establishment/release section 52 A, as a function corresponding with the electrical path establishing section 62 A, and realizes a function corresponding with the virtual routing execution section 65 A which performs virtual routing via the electrical paths established by the electrical path establishment function, and a function corresponding with the congested optical path detection section 54 A which detects the congested locations on the optical paths after this virtual bypassing is performed based on the results of the virtual routing performed by the virtual routing execution section 65 A, and realizes a new optical path establishment function which when a congested optical path as detected by this congested optical path detection section 54 A does not directly connect the electrical sub-networks which are causing the congestion, newly establishes an optical path which directly connects these electrical sub-networks, as a function corresponding with the optical path establishment/release section 52 A. Furthermore, this realizes a function corresponding with the optical path usage ratio determining section 55 A which determines whether or not the usage ratio of the optical paths established by the optical path establishment/release section 52 A is equal to or above a threshold value α, and comprises a function corresponding with the multi-hop electrical path detection section 56 A which detects those electrical paths which do not directly connect between electrical sub-networks which travel optical paths which have the usage ratio equal to or greater than the threshold value α based on the results of the optical path usage ratio determining section 55 A, and a function corresponding with the highest traffic electrical path selection section 57 A which selects the electrical path which carries the highest traffic quantity from the electrical paths detected by this multi-hop electrical path detection section 56 A. The program also realizes an optical path adding function which establishes an optical path so that the electrical path selected by the highest traffic electrical path selection section 57 A directly connects the electrical sub-networks as a function corresponding with the optical path establishment/release section 52 A. Furthermore this realizes a function which determines whether or not the usage ratio of the optical paths established by the optical path establishment/release section 52 A is equal to or below a threshold value β, as a function corresponding with the optical path usage ratio determining section 55 A, and realizes a function corresponding with the virtual bypass processing section 58 A which virtually bypasses electrical paths established over optical paths which have the usage ratio equal to or below the threshold value β to other optical paths based on the determination of the determining function, and a function which detects congested locations upon the optical paths after the virtual routing is performed by the virtual bypass processing section 58 A, as a function corresponding with the congested optical path detection section 54 A, and a function which actually releases the optical paths which have the usage ratio equal to or below the threshold value β after the virtual bypassing has actually been performed by the virtual bypass processing section 58 A, when congested optical paths are not detected by the congested optical path detection section 54 A.

In the description of the present embodiment, the function corresponding with the virtual routing execution section 65 A was provided by the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A, 40 A, but this function could also be provided by the photonic border nodes 1 A to 6 A.

Furthermore, in the description of the present embodiment, the function corresponding with the congested optical path detection section 54 A which detects congested locations on the optical paths based on the results of the virtual routing performed by the virtual routing execution section 65 A or based on the virtual bypass processing performed by the virtual bypass processing section 58 A, the function corresponding with the optical path usage ratio determining section 55 A which determines whether or not the usage ratio of the optical paths established by the optical path establishment/release section 52 A is equal to or above a threshold value α, the function corresponding with the multi-hop electrical path detection section 56 A which detects those electrical paths which do not directly connect between electrical sub-networks which travel optical paths which have the usage ratio equal to or greater than the threshold value α based on the results of the optical path usage ratio determining section 55 A, the function corresponding with the highest traffic electrical path selection section 57 A which selects the electrical path with the highest traffic from the electrical paths detected by the multi-hop electrical path detection section 56 A, the function corresponding with the optical path usage ratio determining section 55 A which determines whether or not the usage ratio of the optical paths is equal to or below a threshold value β, and the function corresponding with the virtual bypass processing section 58 A which virtually bypasses electrical paths established over optical paths which have the usage ratio equal to or below the threshold value β to other optical paths based on this determination, were provided by the photonic border nodes 1 A to 6 A, but these functions could also be provided by the electrical border nodes 11 A, 12 A, 21 A, 22 A, 30 A, 32 A, 40 A.

By storing the program of the second embodiment on the storage medium of the present embodiment, the computer can install the program of the present embodiment using this storage medium. Alternatively, the program of the present embodiment may be directly installed on the computer over a network, from a server on which the program of the present embodiment is stored.

Accordingly, by automatically performing the establishment or release of optimal optical paths according the traffic quantity between the electrical sub-networks using an information processing apparatus such as a computer, an optical/electrical path integrated network and a node which can make effective use of network resources without requiring the intervention of the network administrator can be realized.

The second embodiment is described in further detail below.

All of the electrical sub-networks S 1 to S 4 are interconnected by electrical paths (E-LSPs), in a full mesh. The E-LSPs are routed over the entire electrical network constructed from the electrical sub-networks S 1 to S 4 , which are interconnected by the optical paths (O-LSP) established over the photonic core network C. When the electrical sub-networks S 1 to S 4 are connected by multiple-hop routing, the E-LSP which connects the sub-networks traverses a plurality of O-LSPs.

The traffic quantity between the electrical sub-networks S 1 to S 4 can be determined by counting the packets flowing over the E-LSPs. The traffic between all of the electrical sub-networks S 1 to S 4 can be expressed as a matrix, called a traffic matrix. A traffic matrix is shown in FIG. 18 . The example in FIG. 18 shows the traffic matrix of a network comprising N electrical sub-networks, and the components (i, j) in the matrix represent the traffic quantity from electrical sub-network i to electrical sub-network j.

This traffic matrix is applied to each of the photonic border nodes 1 A to 6 A. The photonic border nodes 1 A to 6 A establish or release optical paths in an autonomous manner based on this information.

The steps involved in establishing O-LSPs based on the traffic matrix are described below. In the present invention, first O-LSPs are established until the electrical sub-networks S 1 to S 4 are “connected”, to use the terminology of graph theory (this is called the connection phase), and the E-LSPs are then routed over the entire electrical network formed by being interconnected by the O-LSPs, and steps to verify whether or not the necessary number of O-LSPs have been established are followed (this is called the capacity verification phase). When establishing the O-LSPs, the shortest paths are calculated based on a topology in which the photonic core network C has free resources, and the O-LSPs are established along these paths.

FIG. 14 shows a flowchart of the connection phase. The E-LSP with the highest traffic quantity is selected from the traffic matrix, and an O-LSP is established so that this E-LSP can be connected by a single hop. If an O-LSP cannot be established at the time, the E-LSP is then marked as having been checked. The E-LSPs are then checked in descending order of traffic quantity, and these steps are repeated until all of the electrical sub-networks are “connected”.

FIG. 15 shows a flowchart of the capacity verification phase. Once all of the electrical sub-networks S 1 to S 4 are “connected”, capacity verification is performed to verify whether sufficient bandwidth for the E-LSP is secured. Virtual routing is then performed, supposing an electrical network in which all of the E-LSPs are interconnected by O-LSPs. In this case, routing is performed in accordance with shortest paths. The shortest paths for the E-LSPs can be calculated based on the topology information of the established O-LSPs and the electrical network.

After virtual routing is performed for all of the E-LSPs, the O-LSPs are checked for congestion. A determination is made that congestion is present if the usage ratio of the O-LSP exceeds the threshold value α. Whether or not the E-LSP with the highest traffic quantity of the E-LSPs which is routed over the congested O-LSP are routed by multiple hop routing over a congested O-LSP can be connected by a direct O-LSP is tested. If such an O-LSP cannot be connected, the E-LSPs are checked in descending order of traffic quantity. Once an O-LSP can be established, virtual routing of all of the E-LSP is performed again, and whether or not the capacity of the O-LSP is sufficient is verified. The steps above are repeated until congestion does not occur in any of the O-LSPs.

FIG. 19 is an E-LSP hop count matrix. This matrix stores the number of hops of the O-LSP via which each E-LSP is routed. This E-LSP hop count matrix is used during the capacity verification phase to search for E-LSPs which are candidates for the establishment of direct O-LSPs. In other words, sometimes an increase in the hop count of the E-LSP leads to an increase in the use of O-LSP resources, and consequently when an O-LSP is congested, it is desirable to reduce the hop count of such E-LSPs.

FIG. 20 shows how an O-LSP is routed over a shortest path over the photonic core network C. By performing routing over the shortest path, it is possible to keep the use of the resources of the photonic core network C to a minimum.

If the traffic matrix changes after the establishment of the O-LSPs has been completed, the O-LSPs need to be reconfigured accordingly. FIG. 21 and FIG. 22 are flowcharts showing the establishment and release of O-LSPs.

FIG. 21 is a flowchart showing how O-LSPs are newly established when the traffic matrix has changed. If the usage ratio of the O-LSP exceeds the threshold value α, the E-LSP which travels by multiple hop routing and has the highest traffic quantity of those passing through the O-LSP is selected, and whether or not an O-LSP which can carry this E-LSP directly via a single hop can be established is checked. If such an O-LSP cannot be established, the E-LSP with the next highest traffic quantity is nominated.

FIG. 22 is a flowchart showing how O-LSPs are released when the traffic matrix has changed. If the usage ratio of an O-LSP is equal to or below a threshold value β, checks are performed to determine whether or not the O-LSP can be released. In this case, checks are performed to determine whether the network would remain “connected” and whether there would be sufficient capacity if the O-LSP is released. To determine whether or not there is sufficient capacity, whether or not another O-LSP has sufficient resources to accept the bandwidth of the E-LSPs which are bypassed after the O-LSP is released is verified. This is achieved by virtually bypassing all of the E-LSPs established over the O-LSP which is to be released to another O-LSP, and determining whether or not congestion occurs on the O-LSP as a result. If it is safe to release the O-LSP, then first the topology after the O-LSP is released is assumed, and the E-LSPs are bypassed in advance. Once all of the E-LSPs have been bypassed, the O-LSP is released.