JP3049301B2 - Failure recovery and congestion recovery in connection-oriented communication networks - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system in which a plurality of nodes are connected by physical links.
In a connection-type communication network in which connections that form a communication path via one or more physical links are accommodated, a failure recovery method when a physical link fails, and when a congestion occurs in this connection The present invention relates to a congestion recovery method, and more particularly, to a high-speed recovery process, a response to a large-scale communication network, and a failure recovery method and a congestion recovery method in a connection-type communication network resistant to failure.

[0002]

2. Description of the Related Art In a communication network composed of a plurality of nodes, as a control method when a failure, congestion, traffic fluctuation, or the like occurs and the communication network needs to be reconfigured, an active / standby network resource is used. The fault / congestion recovery method aiming at maintaining as many connections as possible is to redistribute the connection among the connections, and the "failure VP rescue method considering traffic" (1995 IEICE Communication Society Conference; " Information Network Trust Measures Subcommittee) and Dependable Network
King Service: High reliability of end-to-end service by cooperation of terminal and network "(The Institute of Electronics, Information and Communication Engineers 1996;" Multi-network quality and issues "subcommittee). Then, if the spare resources are sufficient, the fault / congestion recovery is performed by using a recovery technique developed conventionally such as a self-healing method, and if the spare resources are insufficient, the current network is used. It has been proposed to recover a fault / congestion by using a resource with a low priority among resources as a spare.

[0003] As a specific method of using the current network resources as a backup, a "path switching control method considering the connection quality of the line" (February 1993, IEICE Transactions B-
I, Vol. 76, pp. 100-107) ("Intelligent Pa
th Assignment Control ”, IEICE Transaction on Com
munications, BI, J76-No. 2, pp100-107, 1993), a hybrid failure recovery system has been proposed.

In the proposed hybrid system, referring to FIG. 38, when a failure or congestion occurs in a communication network formed by nodes 91 to 94 mutually connected by a physical link 95, the centralized control center 90 The additional switching processing unit optimizes the entire communication network.

That is, the additional switching processing unit of the central control center 90 performs the minimum recovery in a distributed manner based on the switching pattern registered in advance to secure the high speed of the recovery from the failure, and then executes the traffic data and the failure. An additional switching pattern is created based on the data, and the failure data is sent to the nodes 91 to 94 to perform the optimal switching.

In the configuration example of the communication network shown in FIG. 38, centralized control center 90 is a band optimizing server that performs band optimizing by centralized control.
Reference numeral 4 handles the connections that make up the communication path, and processes the formation of the connections. When a failure occurs, each of the nodes 91 to 94 sends failure information to the centralized control center 90 of the bandwidth optimizing server, and the centralized control center 90 of the bandwidth optimizing server calculates an optimal network configuration based on the information. The nodes 91 to 94 are controlled to optimize the entire communication network.

[0007]

In the above-described conventional fault recovery method and congestion recovery method in a connection-type communication network, control using the current network resources as a spare for recovery from a fault / congestion is performed by centralized control. I was doing.

[0008] When recovery control for failure / congestion of the entire communication network is performed by centralized control, the following various problems unique to centralized control occur.

[0009] The first problem is that it is difficult to increase the speed.

[0010] The reason is that, when a failure or congestion occurs, the centralized control unit collects information from each node in order to grasp the state of the communication network, determines a new communication network state, and determines the result. Control each node based on
This is because the amount of information collected from each node and the amount of information transmitted for controlling each node are large, and it is difficult to perform control at high speed. This problem is particularly serious in fault recovery where high speed control is strongly required.

A second problem is that it is difficult to cope with an increase in the scale of a communication network.

The reason is that since the information of the entire communication network needs to be stored in the central control unit, the amount of information increases when the communication network becomes large-scale. Therefore, it is necessary to store the information in one place. This is because it becomes difficult. Furthermore, since the number of processes for performing control increases, it becomes difficult to perform control at high speed.

A third problem is that it is vulnerable to failure.

The reason is that if a failure occurs in the central control unit or the control information communication path connecting the central control unit and each node, the control becomes impossible. Therefore, it is necessary to take measures such as duplication of the centralized control unit and the communication path connecting the centralized control unit and each node so as to reduce the probability of the occurrence of communication interruption due to a failure.

An object of the present invention is to solve the above problems,
An object of the present invention is to provide a high-speed recovery process, a response to a large-scale communication network, and a failure recovery method and a congestion recovery method in a connection-type communication network that is resistant to a failure.

[0016]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, a fault recovery system and a congestion recovery system in a connection-type communication network according to the present invention perform control using distributed network resources in a distributed manner. I have.

In the first aspect, a plurality of nodes are connected by physical links, and each physical link accommodates a connection forming a communication path via at least one physical link. In the failure recovery method when a failure occurs in the physical link,
Each node has a resource management table for managing the setting of a connection passing through its own node. A node that detects the occurrence of a failure on a physical link connected to its own node sends a resource delivery request message containing location information of the failure. The node sends the resource delivery request message to all adjacent nodes if the received resource delivery request message is the first message relating to the failure. The node that has transmitted the resource supply request message changes the connection setting so that more connections can be accommodated in the physical link with respect to the connection in which the failure has not been detected among the connections from which the own node is the transmission source, Connection settings including the settings changed to the next via node of the connection Further request message is sent, the node that received the connection setting change request message,
If the own node is a relay node for the connection, the setting of the resource management table for the connection is changed based on the received connection setting change request message, and the connection setting change request message is sent to the next node. If the node is the destination node for the connection, change the settings of the resource management table for the connection based on the connection setting change request message. If the node is the source of the connection, If the detected connection is set by trying a different path in a distributed manner by communication between the nodes by communication between the nodes before the occurrence of the failure, and the connection setting needs to be changed in order to use more network resources, Determined in advance after receiving the resource delivery request message Is characterized by attempting to set up change of the connection since after a certain period of time has elapsed that was.

The second invention is different from the first invention in that each node further comprises a resource supply flag setting area indicating that the resource supply processing of the connection is being performed, and The node that detects the occurrence of the failure on the physical link connected to the node sends a resource supply request message including the location information of the failure to all the adjacent nodes when the valid resource supply flag is not set. Similarly, the node that has received the resource supply request message sends the received resource supply request message to all adjacent nodes when the valid resource supply flag is not set.

The third aspect of the present invention is different from the first aspect in that a node that detects the occurrence of a failure in a physical link connected to its own node transmits a resource supply request message including location information of the failure. Is sent to all adjacent nodes, and the node that received the resource offer request message sends this resource offer request message to all adjacent nodes if the received resource offer request message is the first message regarding the failure. Then, the node that sent this resource supply request message changes the setting so that more connections can be accommodated in the physical link among all connections from which the own node is the transmission source and for which no failure has been detected. When all of the connection settings have been changed, the completion of all connection setting changes The node that sends the message to all nodes adjacent to its own node, and receives the all connection setting change completion message, if the original transmission source node of the received all connection setting change completion message is the first node, receives all the connection setting change completion messages. That is, a connection setting change completion message is sent to all adjacent nodes.

The fourth invention is different from the third invention in that each node further includes a resource supply flag setting area and detects the occurrence of a failure on a physical link connected to the own node. When the valid resource supply flag is not set, the node that has transmitted the resource supply request message including the location information of the failure to all the adjacent nodes, and the node that has received the resource supply request message also has a valid resource supply request message. If the resource supply flag is not set, it means that the received resource supply request message is sent to all adjacent nodes.

[0021] In a fifth aspect based on the first to fourth aspects, the node performs an attempt to set up a connection by another path when a connection from which the node is a transmission source has failed. It is characterized in that it is performed after a certain period of time has passed since the detection of the occurrence of a connection failure.

In a sixth aspect based on the first to fourth aspects, the node performs an attempt to set up a connection by another path when the connection from which the node is a transmission source has failed. It is characterized in that it is repeated a predetermined number of times.

In a seventh aspect based on the first to fourth aspects, the resource management table of the node includes a resource amount currently used by a connection passing through the own node, a resource amount used at normal times, and a fault amount. Resources used at the time of occurrence,
An area for storing information for identifying a previous node through which a connection passes, information for identifying a next node, and an identifier of an alternative connection is provided, and the bandwidth of each connection is changed when a resource supply request message is transmitted / received. It is characterized in that the resource management table is changed to the resource amount at the time of failure.

In the eighth invention, a plurality of nodes are connected by physical links, and each physical link accommodates a connection forming a communication path via one or more physical links. In the congestion recovery method when congestion occurs in this connection in the communication network, each node has a resource management table that manages the setting of the connection passing through the own node, and detects the occurrence of congestion of the connection connected to the own node The node that has sent the resource delivery request message to all neighboring nodes, and the node that received this resource delivery request message sends it to all neighboring nodes if the resource delivery request message received for the congestion is the first message. The node that has transmitted this resource supply request message, and has transmitted this resource supply request message, For the connection from which the node is the source, change the setting so that more connections can be accommodated on the physical link, and send a connection setting change request message for transmitting the changed setting contents to the next transit node following the connection, If the own node is a relay node for the connection, the node that has received the connection setting change request message changes the setting of the resource management table for the connection based on the connection setting change request message, and transmits the connection setting change request message to the next node. And if the own node is the destination node for the connection, the setting of the resource management table for the connection is changed based on the connection setting change request message. Attempt to increase the bandwidth for the first detected connection, and if it is necessary to change the connection setting to use more network resources, a predetermined setting is required after receiving the resource supply request message. Attempting to change the connection settings after the time has elapsed.

In the ninth aspect, the different point from the eighth aspect is that each node further includes a setting area of a resource supply flag, and the node which detects the occurrence of congestion in the connection connected to its own node If the valid resource supply flag is not set, the resource supply request message is sent to all adjacent nodes. Similarly, if the valid resource supply flag is not set, the node receiving this resource supply request message also transmits the resource supply request message. Sending a resource supply request message to all adjacent nodes is mentioned.

[0026] In a tenth aspect based on the eighth or ninth aspect, the node makes an attempt to increase the bandwidth of the connection performed when the connection from which the node is the transmission source is congested. It is characterized in that it is performed after a certain period of time has passed after detecting the occurrence of congestion.

In an eleventh aspect based on the eighth or ninth aspect, the node sets in advance an attempt to increase the bandwidth of the connection to be performed when the connection from which the node is the transmission source is congested. It is characterized by repeating it a number of times.

In the above-described means according to the present invention, the active network resources are used for recovery control of a failure or congestion. Therefore, a node which has detected that a failure or congestion has occurred requests all nodes to detect a working resource and makes a request. The receiving node supplies the resources of the connection from which it is the source. For this reason, it is possible to perform recovery control of a failure or congestion using active resources without particularly providing a center for performing centralized control.

In the first aspect of the present invention, the resource supply request message is generated and transmitted from the node at the end of the failed link, thereby starting the resource supply processing with the number of messages on the order of the number of nodes and securing the high speed of the resource supply processing. I have. In addition, after the resource providing process is performed at the transmission source node, if there is a standardized connection setting procedure or connection setting change procedure, if there is a standardized connection setting procedure or connection setting change procedure, it is used to notify the relevant node of the resource provision result for each connection. It is possible to

In the second, fourth and ninth inventions, when generating and transmitting the resource supply request message, the generation and transmission are controlled with reference to the resource supply flag, whereby the information is provided for each failure or congestion. And simplifying node configuration and message format. Further, the second invention and the ninth invention are the same with respect to the resource supply processing, and can easily integrate the failure recovery control and the congestion recovery control.

In the third aspect of the present invention, the resource delivery request message is generated and transmitted from the node at the end of the failed link, thereby starting the resource delivery processing with the number of messages on the order of the number of nodes, thereby ensuring the high-speed processing of the resource delivery. doing. Also,
After the resource supply processing is performed at the transmission source node, completion notification is performed for each node for which the resource supply processing has been completed, thereby enabling notification of resource supply completion based on the number of messages on the order of the number of nodes. Usually, since the number of nodes is smaller than the number of connections, the number of messages is reduced as compared with the case of notifying the resource delivery in connection units, and the notification of the completion of the resource delivery is speeded up.

According to the fifth and tenth aspects of the present invention, after a lapse of a predetermined time from the start of the resource supply processing, a detour search is performed when a failure occurs, and a bandwidth is increased when congestion occurs. Then, after at least part of the resource supply processing is completed, an attempt is made to search for a detour or increase the bandwidth, and as a result, the probability of the attempt being successful is increased.

Further, as in the sixth and eleventh inventions, when the retry is not performed even if the message fails, the number of messages for processing such as a detour search or an increase in the bandwidth is small, so that these processings are speeded up. You.

Further, in the sixth and eleventh inventions, when a failure occurs, a detour search is performed, and when congestion occurs, a bandwidth increase is attempted a plurality of times, so that the resource supply processing is not always completed. Can also initiate a detour search or bandwidth increase attempt. As a result, when the time for completing the resource supply processing is not constant, the resources for which the resource supply has been completed can be used for detour search or bandwidth increase. This is particularly effective when a long time is required for the resource supply processing.

According to the seventh aspect of the present invention, when a resource supply request message is received, the amount of resources used by each connection during normal operation and during a failure is set in advance and stored in the resource management table. The amount of change in the bandwidth of the connection can be determined at high speed, and high-speed resource supply processing is ensured.

According to the eighth aspect, the resource delivery request message is generated and transmitted from the node at the congestion-occurring connection end, thereby starting the resource delivery process with the number of messages in the order of the number of nodes, thereby increasing the speed of the resource delivery process. Secure. In addition, after the source node performs resource supply processing,
If there is a standardized connection setting procedure or connection setting change procedure, it is possible to use the standardized connection setting procedure or connection setting change procedure in order to transmit the result of resource provision to the relevant node for each connection. In addition, the present invention and the first invention,
The same applies to the resource supply processing, and it is possible to easily integrate the failure recovery control and the congestion recovery control.

[0037]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of a failure recovery system and a congestion recovery system in a connection type communication network according to the present invention will be described with reference to the drawings.

The difference of the system configuration of the present invention from the conventional one is that, as described in the above-mentioned "Means for Solving the Problems", the control that uses the current network resources of the communication network in which the nodes are connected by the physical link as a spare. Is not distributed but performed in a decentralized manner. Therefore, each node has its own recovery means for failure and congestion.

In the following description, various notifications are used for information transmission within the own node, and various messages are used for information exchange with other nodes.

First, a first embodiment of the present invention will be described with reference to FIGS.

FIG. 1 shows a node that performs a failure recovery in the first embodiment of the present invention. This node is an interface unit (hereinafter, abbreviated as an IF unit).
1, a resource control unit 2, a switch unit 3, a central control unit 4, a connection acceptability determining unit 5, a timer 6, a resource supply event management table 7, and a resource management table 8, and the connection acceptability determining unit 5 In addition to connecting to the management table 8, the central control unit 4 connects to each component to exchange information.

The nodes are connected by a physical link 11, and a connection 12 as a communication path is accommodated in the physical link 11. Specific examples of the connection 12 include a path or a line, a virtual path or a virtual channel.

The IF unit 1 monitors the occurrence of a failure in the physical link 11 or the connection 12. Also, I
The F unit 1 extracts a message sent from another node, transmits the message to the central control unit 4, and inserts the message generated by the central control unit 4 into the connection 12.

The resource control unit 2 sends and receives information to and from the central control unit 4 to monitor the bandwidth of the connection 12. Details will be described later with reference to flowcharts.

The switch section 3 cross-connects the connection 12 under the control of the central control section 4.

The central control unit 4 has band management means and network control means (not shown), receives various notifications from the IF unit 1 together with messages from other nodes, and controls the resource control unit 2 and the switch unit 3. Change the settings by controlling
The entire node is controlled, including sending a message to another node via the unit 1. The main functional operation of the central control unit 4, which exchanges information with each component and controls the entire node, will be specifically described later with reference to flowcharts.

The connection acceptability judging section 5 exchanges information with the resource management table 8 under the control of the central control section 4 to judge the acceptability of the connection 12.

The timer 6 measures time under the control of the central control unit 4, and notifies the central control unit 4 of a predetermined notification when the designated time has elapsed.

The resource supply event management table 7 is controlled by the central control unit 4, and is managed by recording and updating the event occurrence status at that time.

The resource management table 8 records and stores the current bandwidth use status of the connection 12, and the recorded bandwidth use status is referred to by the central control unit 4 and the connection acceptability determination unit 5.

In the first embodiment, when a failure occurs, a detour path is set in advance by setting an alternative connection having a bandwidth of “zero” for each connection, and changing the bandwidth of the connection to a failure. It is assumed that a method of setting by changing at some time is used. The setting change of the connection so that the physical link can accommodate more connections is performed by reducing the bandwidth of the already set connection.

Next, FIG. 2 shows an example of information contained in various notifications and various messages used in the first embodiment of FIG.

As shown in FIG. 2, the notice includes a link failure occurrence notice, a connection failure occurrence notice, a connection switching request notice, and an all connection resource acquisition permission notice. , And the failure occurrence time, the connection failure occurrence notification and the connection switching request notification respectively include the message type and the connection identifier, and the all connection resource acquisition permission notification includes only the message type.
Each shall be included.

The messages include a resource supply request message, a connection setting change request message, a connection setting change rejection message, a connection switching request message, and a connection switching rejection message. The resource supply request message includes a message type and an event type. , The event occurrence time, and the event occurrence location, the remaining connection setting change request message, connection setting change rejection message, connection switching request message, and connection switching rejection message, message type, connection identifier, and bandwidth change width, respectively Shall be included.

Next, FIG. 3 shows an example of the format of the resource supply event management table 7 in FIG.

In this format, a physical link in which a failure has occurred is specified by specifying an identifier of a node and a port. In addition, it is also possible to specify a physical link by a method of giving a unique identifier to each physical link in the entire communication network, or specifying nodes on both sides.

In the resource supply event management table 7 shown in FIG. 3, a physical link failure occurs at 10:08:05 on the port 3 of the node 5 corresponding to the location and the time for the event type. , Connection congestion at port 6 of node 12, line number 220 at 10:08
It is shown that it occurred at minute 12 seconds.

Next, FIG. 4 shows an example of the format of the resource management table 8 in FIG. Note that the area of the alternative connection identifier is valid only in the case of the originating node or the destination node.

In the resource management table 8 shown in FIG.
10Mb bandwidth corresponding to connection identifier "1"
ps, previous transit node identifier / link identifier “none”, next transit node identifier / link identifier / link width “5/12
/ 150 Mbps ", source node" yes ", destination node" no ", and alternative connection identifier" 18 "
, And corresponding to each of the connection identifiers “2, 3,...”, The bandwidth is 64 Kbps, 6.3 Mbps,.
Previous via node identifier / link identifier "8/10, 8/1
0 ", next transit node identifier / link identifier / link width" none, 5/12/150 Mbps, ... ", originating node" no, no, ... ", destination node" yes, no,
… ”And the alternative connection identifier“ 20, none,
.. Are recorded and stored.

Next, the processing operation of the basic function of the first embodiment will be described with reference to FIG. 5 and FIGS.

First, in step 501, the IF unit 1 detects the occurrence of a failure in the physical link 11 connected to the own node, and sends a link failure notification to the central control unit 4. As shown in FIG. 2, the link failure notification has, as additional information, a pair of a node identifier and a port identifier indicating a location where a failure has occurred, and a failure occurrence time.

In step 502, the central control unit 4 having received the notification of the occurrence of the link failure records in the resource supply event management table 7 that an event to start resource supply has occurred.

Next, in step 503, the central control unit 4
Sends a resource delivery request message to all adjacent nodes to start resource delivery. The resource supply request message also includes information necessary for creating the resource supply event management table 7.

When each physical link is specified by a set of a node identifier and a port identifier similarly to the case of the resource supply event management table 7 shown in FIG. 3, the resource supply request message shown in FIG. , The time of the link failure occurrence notification is specified as the event occurrence time, and a set of a node identifier and a port identifier is specified as the event occurrence location. When a unique identifier is given to each physical link in the entire communication network, the physical link identifier is specified at the event occurrence location.

Next, in step 504, the central control unit 4
Checks whether there is a connection in which the own node is the originating node by using the resource management table 8.

If the procedure 505 is “YES” and there is a connection in which the own node is the originating node, the resource delivery processing from the procedure 506 to the procedure 509 is performed. Reduce the bandwidth of the connection. On the other hand, if the procedure 505 is “NO” and there is no connection whose own node is the originating node, the processing from the procedure 506 to the procedure 509 is skipped.

The process of reducing the connection bandwidth is started in step 506. That is, the central control unit 4
Controls the resource control unit 2 to reduce the bandwidth of the target connection, and stores the reduced result in the resource management table 8.
Set to.

The reduction width of the bandwidth may be determined in advance, or at the time of the reduction processing, the utilization rate of the connection, the location of the failure, the number of the failures, and the physical link of the failed physical link. It may be determined according to the bandwidth or the like. When it is determined at that time that it is not preferable to reduce the bandwidth of the connection, the bandwidth need not necessarily be reduced. Also, this determination method may be different for each node.

Next, in step 507, the central control unit 4
Sends a connection setting change request message to the next transit node of the connection whose bandwidth has been reduced.

The connection setting change request message is
As shown in FIG. 2, the information includes the identifier of the connection whose bandwidth is to be changed and the change width. This change width may be the bandwidth after the change.

Finally, in step 508, the central control unit 4 checks whether or not the bandwidth reduction processing procedure for all connections originating from the own node has been completed.
If "O" and the bandwidth reduction processing procedure for all connections has not been completed, the procedure returns to step 506, and the above procedure is repeated.

If the procedure 509 is “YES” and the procedure for reducing the bandwidth of all connections is completed by the procedure so far, the central control unit 4 sets the time Tgr in the timer 6 in the procedure 510 and ends the procedure. . When the time Tgr has timed out, the timer 6 generates an all connection resource acquisition permission notification and sends it to the central control unit 4.

This connection resource acquisition permission notice is:
As shown in FIG. 2, there is no special additional information. This time-out is provided for a failed connection to perform a fault detour using an unused bandwidth caused by a decrease in the bandwidth of an existing connection by the processing up to this point. Therefore, the time Tgr is set to a time enough to complete the bypass setting of a sufficient number of failed connections.

Next, with reference to FIG. 6 and FIGS. 1 to 4, a description will be given of the processing when the IF unit 1 receives the resource supply request message transmitted in step 503 of FIG. 5 from another node. .

First, in step 601, the IF unit 1 receives a resource supply request message sent from another node. As shown in FIG. 2, the resource supply request message has, as additional information, information for specifying the time of occurrence of the failure and the link in which the failure has occurred.

In step 602, the central control unit 4 refers to the resource management table 8 and checks whether the same resource supply request message relating to this failure has already been received.

If step 602 is “YES” and the same resource supply request message has already been received, steps 603 to 610 are skipped, and the procedure proceeds to step 611.

If the procedure 602 is “NO” and the same resource supply request message has not yet been received, the central control unit 4
Records the received failure event occurrence in the resource supply event management table 7 in step 603, and sends a resource supply request message to all the adjacent nodes in step 604 with the same additional information as the received information. .

Next, in step 605, the central control unit 30
2 uses the resource management table 8 to check whether or not there is a connection whose own node is the originating node.

If “YES” is determined in step 606 and there is a connection in which the own node is the originating node, the central control unit 4 determines in step 607 to step 610
Processing to reduce the bandwidth of those connections is performed.

On the other hand, if “NO” in the step 606 and there is no connection in which the own node is the originating node, the central control unit 4 skips the processing from the steps 607 to 610 and proceeds to the following step 611.

In the process of reducing the bandwidth of the connection, first in step 607, the central control unit 4 controls the resource control unit 2 to reduce the bandwidth of the target connection, and based on the result, the resource management table Change the settings of 8.

The width of the decrease in the band may be determined in advance, or in the process of decreasing the band, the resource supply event management table 7 and the use state of the connection 12 are referred to and determined in accordance with the reference state. Is also good. For example, if a failure occurs on a link that must pass through many links from the own node, the amount of bandwidth reduction is reduced, and the failure of the plurality of physical links 11 is determined by the resource delivery event management table 7.
When recorded in the
When a failure occurs in the large-capacity physical link 11, a method of increasing the amount of bandwidth reduction can be considered.

Next, at step 608, the central control unit 4
Sends a connection setting change request message to the next transit node of the target connection. As shown in FIG. 2, the connection setting change request message has, as information, the identifier of the target connection for which the band is to be changed and the band change width.

Next, in step 609, the central control unit 4
Examines whether the bandwidth reduction of all the connections originating from the own node has been completed. If the procedure 610 is “NO” and the bandwidth reduction of all the connections has not been completed yet, the procedure from the procedure 607 to the procedure 610 is completed. Repeat the procedure.

On the other hand, if the procedure 610 is “YES” and the bandwidth reduction of all connections is completed, or if the procedure 606 is “NO” and there is no connection to become the initial node, The control unit 4 sets the time Tgr in the timer, and receives a notification of permission to acquire all connection resources generated when a timeout occurs, and ends the processing. This connection resource acquisition permission notification does not have any special additional information.

The processing from step 603 to step 611 corresponds to step 502 in FIG. 5 when receiving a link failure notification.
To the process 510.

Next, referring to FIG. 7 and FIGS. 1 to 4, the connection setting change request message transmitted in step 507 in FIG. 5 or step 608 in FIG.
Will be described in the case where is received from another node.

First, in step 701, the IF unit 1 receives a connection setting change request message sent from another node. As shown in FIG. 2, the connection setting change request message has an identifier of the target connection and a band change width as additional information.

In step 702, the central control unit 4 checks whether or not the own node is the destination node for the connection that has received the bandwidth change request. Step 702 is “YES”
In step 703, the central control unit 302 changes the setting of the resource management table 8 and ends the process.

On the other hand, if the procedure 702 is “NO” and it is not the destination node,
When the bandwidth of the specified connection is changed by the specified width, the connection acceptability judging section 5 checks whether or not the next connection physical link can be accommodated.

If “YES” is determined in step 705 and the next pass physical link can be accommodated, the central control unit 4 proceeds to step 7
In step 06, the bandwidth of the target connection is changed by the resource control unit 2 to change the setting of the corresponding portion of the resource management table 8, and in the next step 707, a connection setting change request message is sent to the next transit node of the target connection. Is sent to end the procedure.

If the connection identifier of the connection setting change request message is unique within the communication network,
If the identifier is the same as the received connection setting change request message and is unique only for each node and the identifier is converted each time the node passes, the converted identifier is used. The bandwidth change width is the same as the received connection setting change request message.

On the other hand, if it is determined in step 705 that the data cannot be accommodated in the next transit physical link in step 705, the central control unit 4 sends the connection request message in step 708 to the change request source node that sent the connection setting change request message. A connection setting change rejection message is transmitted via the via node, and the procedure ends.

The connection identifier of the connection setting change rejection message is the same as the received connection setting change request message, and the bandwidth change width is set to a value obtained by reversing the sign of the connection setting change request message.

Next, with reference to FIG. 8 and FIGS. 1 to 4, the processing when the IF unit 1 receives the connection setting change rejection message transmitted in step 708 of FIG. 7 from another node. explain.

First, in step 801, the IF unit 1 receives a connection setting change rejection message transmitted from another node. As shown in FIG. 2, the connection setting change rejection message has, as additional information, an identifier of the target connection and a bandwidth change width for restoring the bandwidth.

In step 802, the central control unit 4 receiving the connection setting change rejection message returns the bandwidth of the target connection by the resource control unit 2,
The setting of the corresponding part is changed in the resource management table 8.

Next, in step 803, the central control unit 4
Determines whether the own node is the destination node for the target connection. If the procedure 803 is “YES” and it is the destination node, the procedure ends.

On the other hand, if the procedure 803 is “NO” and it is not the destination node,
Sends a connection setting change rejection message to the change request source via the via node before the connection,
End the procedure.

The bandwidth change width of this connection setting change rejection message is the same as the received connection setting change rejection message. The connection identifier is the same as the received connection setting change rejection message when it is unique within the communication network, and is changed accordingly when it changes for each node, as in the case of receiving the connection setting change request message.

Next, the processing when the IF unit 1 detects a connection failure will be described with reference to FIG. 9 and FIGS.

First, in step 901, the IF unit 1 detects the occurrence of a failure in a connection connected to its own node, and sends a notification of the occurrence of a connection failure to the central control unit 4.
As shown in FIG. 2, the connection failure occurrence notification has the identifier of the failed connection as additional information.

Next, in step 902, the central control unit 4
Determines whether the own node is the originating node of the target connection. If the procedure 902 is “NO” and it is not the originating node, the processing procedure ends.

On the other hand, if the procedure 902 is “YES” and the originating node, the central control unit 4 sets the time Tsh in the timer 6 in the procedure 903, and after the time out of the time Tsh, the central control unit 4 sets the time as the connection identifier as additional information. Processing is performed to receive a connection switching request notification having the failed connection identifier from the timer 6, and the procedure ends.

Next, a process when the central control unit 4 receives a connection switching request notification from the timer 6 will be described with reference to FIG. 10 and FIGS.

First, in step 1001, the central control unit 4 receives a connection switching request notification. As shown in FIG. 2, the connection switching request notification has the identifier of the failed connection as additional information called the connection identifier before switching.

Next, in step 1002, the central control unit 4
The connection acceptability determining unit 5 refers to the resource management table 8 to determine the alternative connection of the failed connection and the next physical link following the connection, and accommodates the next connection physical link even if the bandwidth of the alternative connection is increased. Investigate the possibilities.

Even if the bandwidth of the alternative connection after the switching is equal to the bandwidth of the failed connection set in the resource management table 8, it is equal to 80 times the bandwidth of the failed connection.
% May be determined in advance, or may be determined from the use status of the failed connection or the current traffic status of the communication network when a failure occurs.

If the connection acceptability determining unit 4 determines in step 1003 that the bandwidth cannot be increased, the processing procedure ends.

On the other hand, if the connection acceptability determination unit 4 determines that the bandwidth can be increased by “YES” in the procedure 1003, the central control unit 4 in the procedure 1004 changes the bandwidth management unit to the alternative connection of the failed connection. Is switched and the bandwidth of the alternative connection is increased, and the network control means changes the setting relating to the bandwidth of the alternative connection in the resource management table 8.

Next, in step 1005, the central control unit 4
Sends a connection switching request message to the next transit node of the alternative connection. In the connection switching request message, the identifier of the alternative connection is set as the connection identifier, and the increase width of the bandwidth is set as the bandwidth change width. The processing procedure when the connection switching request notification is received by transmitting the connection switching request message ends.

Next, with reference to FIG. 11 and FIGS. 1 to 4, a description will be given of the processing when the IF unit 1 receives a connection switching request message from another node in the procedure 1005 of FIG.

First, in step 1101, the IF unit 1 receives a connection switching request message sent from another node. As shown in FIG. 2, the connection switching request message has, as additional information, an identifier of a connection whose bandwidth is to be changed and a bandwidth change width.

Next, in step 1102, the central control unit 4
Investigates the up-to-dateness of whether there is the same bandwidth change request for the connection that has received the bandwidth change request. If “NO” in procedure 1103 and there is the same bandwidth change request, the processing procedure ends.

On the other hand, if “YES” in step 1103 and there is no same band change request, and if it is the first band change request, in next step 1104, the central control unit 4 determines whether the connection receiving the band change request Is a destination node.

If “YES” is determined in step 1104 and the own node is the destination node, in step 1105 the central control unit 4 checks which connection has received the bandwidth change request as an alternative connection. The connection is switched by the resource control unit 2 based on the information, and the setting of the bandwidth of the connection that has received the bandwidth change request in the resource management table 8 is changed, and the processing procedure ends.

On the other hand, if “NO” in step 1104 and the own node is not the destination node, in step 1106 the central control unit 4 sets the connection bandwidth specified by the connection acceptability determination unit 5 to the specified width. In the case of a change, check the margin to see if it can be accommodated in the next transit physical link.

In the step 1107, if “YES” in the step 1107, the central control unit 4 can be accommodated in the next-passing physical link and the connection band can be changed by the specified width, the central control unit 4 is targeted by the resource control unit 2 in the step 1108. The bandwidth of the connection is changed, and the setting of the corresponding part is changed in the resource management table 8.

Next, in step 1109, the central control unit 4
Sends a connection switching request message to the next via node following the target connection, and ends the processing procedure. At this time, the additional information stored in the connection switching request message is the same as the received connection switching request message when the connection identifier is unique within the communication network. If the connection identifier changes for each node, the converted identifier is set for the connection identifier.

On the other hand, “NO” in the above procedure 1107
If it is determined that cannot be accommodated in the next physical link and the bandwidth of the connection cannot be changed by the specified width,
In step 1110, the central control unit 4 sends a connection switching rejection message to the requesting node that sent the connection switching request message via the previous transit node, and ends the processing procedure. The connection identifier set in the connection switching rejection message is the same as the received connection switching request message. The bandwidth change width is set by reversing the sign of the received connection switching request message.

Next, with reference to FIG. 12 and FIGS. 1 to 4, a description will be given of the processing when the IF unit 1 receives a connection switching rejection message from another node in the procedure 804 of FIG.

First, in step 1201, the IF unit 1 receives a connection switching rejection message transmitted from another node. As shown in FIG. 2, the connection switching rejection message has, as additional information, the identifier of the connection whose bandwidth change was rejected and the bandwidth change width for restoring the bandwidth.

Upon receiving the connection switching rejection message, the central control unit 4 returns the bandwidth of the target connection by the resource control unit 2 in step 1202, and changes the setting of the corresponding part in the resource management table 8.

Next, in step 1203, the central control unit 4 determines whether the own node is a destination node for the target connection. In step 1203, “N
If O "is not the destination node, in step 1204, the central control unit 4 sends a connection switching rejection message to the switching request source to the node before the connection, and ends the processing procedure.

The bandwidth change width of the outgoing connection switching rejection message is made equal to the bandwidth of the received connection switching rejection message. The connection identifier is the same as the received message if the connection identifier is unique in the communication network, and is changed accordingly if it changes for each node.

On the other hand, if “YES” in the step 1203 and the destination node is determined, in the step 1205, the central control unit 4 refers to the resource management table 8 to identify the connection before switching, and performs the switch back. After that, the setting of the corresponding part is changed in the resource management table 8, and the processing procedure ends.

Next, with reference to FIG. 13 and FIGS. 1 to 4, a description will be given of the processing in the case where the central control unit 4 has received the all connection resource acquisition permission notification from the timer 6.

First, in step 1301, the central control unit 4 receives from the timer 6 a notification of permission to acquire all connection resources. This all connection resource acquisition permission notice is shown in FIG.
Has no additional information as shown in FIG.

Next, in step 1302, the central control unit 4
Searches the resource management table 8 and checks whether or not there is a connection in which the own node is the originating node and a connection with insufficient bandwidth exists. Step 130
If the connection with insufficient bandwidth does not exist in “NO” in 3, the processing procedure ends.

On the other hand, if there is a connection for which the bandwidth is insufficient in step 1303 (“YES”), the central control unit 4 increases the bandwidth by performing the processing from the following steps 1304 to 1309. Try.

If there is a connection with insufficient bandwidth, the central control unit 4 selects a connection with insufficient bandwidth in step 1304, and increases the bandwidth of the connection so that the connection to the next physical link is established. Whether or not the connection is possible is checked by the connection acceptability determining unit 5.

If it is determined in step 1305 that the physical link can be accommodated in the next pass, the central control unit 4 in step 1306 increases the bandwidth of the target connection in the central control unit 4, and the network control means The setting of the corresponding part is changed in the resource management table 8 based on the result of the change.

Next, in step 1307, the central control unit 4
Sends a connection setting change request message with the identifier of the connection to be changed and the bandwidth change width as additional information to the node through which the target connection passes next.

Next, in step 1308, the central control unit 4
Checks whether all connections with insufficient bandwidth have been processed, and if "NO" in the step 1309, the process for the connection with the insufficient bandwidth remains, returning to the step 1304 and repeating the processing procedure.

[0136] In step 1309, if "YES" indicates that there is no process remaining for the connection with insufficient bandwidth, the processing procedure ends.

If the determination in step 1305 is "NO" and the physical link cannot be accommodated in the next pass-through physical link, the procedure proceeds to step 1308 to investigate the processing of the connection with a lack of bandwidth.

It should be noted that the cause of the shortage of the bandwidth of the connection investigated in the above-mentioned procedure 1302 does not matter. That is, it does not matter whether the bandwidth is insufficient due to the bandwidth reduction processing performed in FIGS. 5 and 6 or the bandwidth is insufficient due to other factors such as traffic fluctuation.

The above is the first embodiment.

In the first embodiment, the bypass connection is set by setting an alternative connection of the band “0” in advance and changing the band of the connection at the time of failure. Various methods that have been conventionally developed can be used as the connection setting method.

For example, in this setting control method, only a route candidate is determined before a failure, and the route setting itself is performed at the time of failure, or a method of providing a plurality of candidates for a route and selecting from among them when a failure occurs. There are a method of exchanging routing information at the time of failure and searching for a route based on the exchanged information, a method of searching for a route using routing information exchanged as needed, and the like. The route search method using the routing information mentioned at the end of the above includes, for example, ATM
(Asynchronous Transfer Mode) In the case of a network, an ATM Forum PNNI S
WG, "PNNI 1.0", Mar. There is a method standardized as "1996".

Further, in the first embodiment, the setting is changed so that more connections can be accommodated in the physical link by reducing the bandwidth of the already set connection. The setting change method for accommodating this connection may be another method.

In a communication network such as an ATM network capable of controlling the quality, it is possible to change the connection setting by lowering the quality of the already set connection. In this case, in addition to the connection bandwidth, the resource management table
It is necessary to store the quality such as the cell loss rate.

FIG. 14 shows an example of a resource management table used for such a communication network. As shown, FIG.
, The cell discard rates “1.0e-9, 1.0e-3, 1.0e-15,” for the connection identifiers “1, 2, 3,.
… ”And delay fluctuation“> 3 msec,> 3
. msec, 500 msec,... ".

The connection setting change request message, the connection setting change rejection message, the connection switching request message, and the connection switching rejection message need to have the quality of the connection as additional information in addition to the bandwidth. Further, upon receiving the notification of permission to acquire all connection resources, an attempt is made to improve the quality of connections of insufficient quality in addition to the connections of insufficient bandwidth.

SDH (Synchronous Digital Hierarchy)
) Even if the bandwidth and quality of the connection is fixed, such as a network, if multiple connections are established between the same destinations, release some of the connections and change the connection settings. Is feasible.

In the first embodiment, the bandwidth of the failed connection is not changed and reused. This is because it is assumed that the connection is restored after the recovery from the failure. The bandwidth is released when processing the connection switching request notification and the connection switching request message, and a bandwidth release message is sent from the originating / destination node along the path of the failed connection, and the bandwidth is reused. The utilization efficiency of the communication network increases.

Next, a second embodiment of the present invention will be described with reference to FIG.
This will be described with reference to FIG.

FIG. 15 shows a node that executes failure recovery in the second embodiment of the present invention. This node includes an IF unit 1, a resource control unit 2, a switch unit 3, a central control unit 24, a connection acceptability determination unit 25, a timer 26, a resource supply flag area 27, and a resource management table 2.
8 and a resource supply completion node table 29. The connection acceptability determining unit 25 is connected to the resource management table 28, and the central control unit 24 is connected to each component to exchange information. .

The nodes are connected by a physical link 11, and a connection 12 is accommodated in the physical link 11. Specific examples of the connection 12 include a path or a line, a virtual path or a virtual channel.

This node has almost the same configuration as the node shown in FIG. 1 in the first embodiment. The difference is that a resource delivery completion node table 29 is added and the resource delivery event management is performed. There is a resource supply flag area 27 instead of the table 7 (FIG. 1). According to this change, the configuration and functions of the central control unit 24, the connection acceptability judging unit 25, the timer 26, and the resource management table 28 At least one has additions and changes.

In FIG. 15, components having the same functions as those of the node shown in FIG. 1 are given the same reference numerals and description thereof is omitted.

The functions of the central control unit 24, the connection acceptability judging unit 25, and the timer 26 will be described later in the procedure with reference to the flowchart.

The resource supply flag set in the resource supply flag area 27 is a flag indicating that the resource supply processing is being performed, and an example thereof will be described later with reference to FIG.
An example of the recorded contents of the resource management table 28 will be described later with reference to FIG. The resource supply completion node table 29
9 is a list for managing generation and transmission of all connection setting change completion messages.

Also in the second embodiment, when a fault occurs, a detour path is set in advance by setting an alternative connection having a bandwidth of “0” for each connection, and setting a bandwidth of this connection to a fault. It is assumed that a method of setting by changing at the time of is used. The setting change of the connection so that the physical link can accommodate more connections is performed by reducing the bandwidth of the already set connection.

Next, FIG. 16 shows various notifications used for information transmission within the own node and various messages used for information exchange with another node used in the second embodiment of FIG. The following shows an example of the information possessed.

In FIG. 16, the notice includes a link failure occurrence notice, a connection failure occurrence notice, a connection switching request notice, and a connection bandwidth increase request notice. The link failure occurrence notice includes a message type, a failure link identifier, and a failure occurrence. It is assumed that the time includes the message type and the connection identifier in each of the connection failure occurrence notification and the connection switching request notification, and the message type, the connection identifier, and the bandwidth change width in the connection bandwidth increase request notification.

The messages include a resource supply request message, a complete connection setting change message, a connection switching request message, a connection switching rejection message, a connection setting change request message, and a connection setting change rejection message. Contains the message type, the originating node identifier, and the serial number in the all connection setting change completion message, the connection switching request message, the connection switching rejection message, the connection setting change request message, and Each connection setting change rejection message has a message type, connection identifier,
It is assumed that the band change width and the serial number are respectively included.

FIG. 17 shows a format of the material supply flag set in the material supply flag area 27 in FIG. In this example, a flag setting time of 13:45:20 and a flag end time of 13:50:20 are set for the flag “on”.

Next, FIG. 18 shows one format of the resource management table 28 in FIG.

In the illustrated resource management table 28, the first
A recording area for a serial number is added to FIG. 4 in the above embodiment, and the bandwidth is increased to three types of current / normal / failure. That is, for each of the connection identifiers “1, 2, 3,.
10M / 6bps, 64K / 64K / 64Kbps,
6.3M / 6.3M / 0Mbps,..., And the serial numbers "3305, 4720, 0035,.

Next, the processing operation of the basic function in the second embodiment will be described with reference to FIG. 19 and FIGS. 15 to 18.

First, in step 1901, the procedure IF unit 1 detects the occurrence of a failure in the physical link 11 connected to its own node, and sends a link failure notification to the central control unit 24. As shown in FIG. 16, the link failure occurrence notification has a link in which a failure has occurred and a failure occurrence time as additional information.

In step 1902, the central control unit 24, having received the notification of the occurrence of the link failure, sets the resource supply flag area 27
, A resource supply flag for requesting a bandwidth reduction effective for a time Tgr from the time when the link failure occurrence notification is received is set. As shown in FIG. 17, the resource supply flag area 27 sets a flag end time at which the time Tgr elapses with respect to the flag setting time.

Next, in step 1903, the central control unit 2
4 sends a resource delivery request message to all adjacent nodes to start resource delivery. The failure occurrence time of the link failure occurrence notification is set in the event occurrence time of the resource supply request message.

Next, in step 1904, the central control unit 24
Checks whether there is a connection 12 whose own node is the originating node using the resource management table 28. If the procedure 1905 is “YES” and there is a connection 12 whose own node is the originating node, the procedure 19
From 06 to procedure 1908, a process of reducing the bandwidth of those connections is performed.

That is, in step 1906, in the central control unit 24, the bandwidth management unit reduces the bandwidth of the target connection, and the network control unit changes the setting of the corresponding portion in the resource management table 28 by the reduced bandwidth. Then, the central control unit 24 checks in step 1907 whether or not the bandwidth reduction of all connections originating from the own node has been completed. In step 1908, “NO” has not been completed for the bandwidth reduction of all connections. If so, step 1906
And the procedure is repeated to reduce the connection bandwidth.

When the above procedure 1905 is “NO” and there is no connection 12 whose own node is the originating node,
Alternatively, if “YES” is determined in the above procedure 1908 and the processing for reducing the bandwidth of the connection is completed for all the connections in which the own node is the originating node, the procedure 190
In step 9, the central control unit 24 of the central control unit 24 sends an all connection setting change completion message to all adjacent nodes, and ends the processing procedure.

In the all connection setting change completion message, the identifier and the serial number of the own node are added as additional information. In the second embodiment, the serial number does not guarantee the time order of arrival of all connection setting change completion messages. Therefore, the connection number such as an all connection setting change completion message, a connection switching request message, a connection setting change request message, etc. When a message for performing a band change notification / band change request is received, the order in which these messages are generated at the connection originating node is determined, and is used to maintain the consistency of the band.

Accordingly, the serial number may be given so as to be unique in a unit in which the all connection setting change completion message is transmitted in the processing of the present link failure notification, that is, in a node unit.

The bandwidth change width of the connection in the above procedure 1906 is set in advance and notified to the related nodes in advance so that the resource management table 28 can be changed only by receiving the all connection setting change completion message. The resource management table 28 in this case has a format as shown in FIG. 18, for example.

Next, a process in the case where the IF unit 1 receives a resource supply request message from another node in the second embodiment will be described with reference to FIGS.

First, in step 2001, the IF unit 1 receives a resource supply request message from another node and sends it to the central control unit 24. The resource delivery request message is shown in FIG.
As shown in FIG. 6, a failure occurrence time is included as additional information.

In step 2002, the central control unit 24
Checks whether a valid resource supply flag is set in the resource supply flag area 27 at the time of receiving the resource supply request message.

If the valid resource supply flag is set to “YES” in step 2003, the central control unit 24 determines in step 2004 that the resource reduction request message received the bandwidth reduction start time of the set flag. Is compared with the band reduction start time. Step 2
If the comparison times are “NO” in 005, the central control unit 24 regards the received resource supply request message as a message relating to the same failure, and has already performed the processing relating to the failure. To end. Or, if the set time is determined to be new,
The central control unit 24 considers that the processing corresponding to the more recent failure has already been performed, and terminates the processing.

On the other hand, if the flag valid at “NO” is not set in step 2003, or if step 20
If "YES" in 05 and only processing relating to a failure older than the message received this time is performed,
In step 2006, the central control unit 24 sets a flag with the failure occurrence time as the band reduction start time.

Next, in step 2007, the central control unit 2
In 4, the network control means sends a resource delivery request message having the same content as the received resource delivery request message to all adjacent nodes. This is a process performed for distributing a resource delivery request message having the same content to the entire communication network.

Next, in step 2008, the central control unit 2
No. 4 uses the resource management table 28 to check whether or not there is a connection whose own node is the originating node. If “YES” in step 2009 and there is a connection in which the own node is the originating node, step 201
At 0, the central control unit 24 performs processing to reduce the bandwidth of those connections.

In step 2010, in the central control unit 24, the bandwidth management means reduces the bandwidth of the target connection, and the network control means changes the setting in the resource management table 28 to the reduced bandwidth.

Next, in step 2011, the central control unit 2
4 examines whether the bandwidth reduction processing of all connections originating from the own node has been completed, and if the procedure 2012 is “NO” and the bandwidth reduction processing of all connections has not been completed, the procedure returns to the procedure 2010 and the procedure returns. I repeat.

On the other hand, in the above procedure 2009, "NO"
If there is no connection in which the own node is the originating node, the central control unit 24 proceeds from step 2010 to step 201.
The process skips the processing procedure up to 2 and proceeds to the procedure 2013.

If the procedure 2009 is “NO” and no connection exists, or if the procedure 2012 is “YES” and the bandwidth reduction processing of all connections is completed, the central control unit 24
The network control unit sends an all connection setting change completion message to all adjacent nodes, and ends the processing.

Note that the processing procedure from step 2007 to step 2013 is the same as the processing procedure from step 1903 to step 1909 in FIG. 19 when a link failure notification is received.

Next, a processing procedure in the case where the IF unit 1 receives an all connection setting change completion message in the second embodiment will be described with reference to FIGS.

First, in step 2101, the IF unit 1 receives an all connection setting change completion message sent from another node and sends it to the central control unit 24. As shown in FIG. 16, the all connection setting change completion message has, as additional information, the node identifier of the node whose bandwidth has been reduced and the serial number.

At step 2102, the central control unit 24
Whether or not the received all connection bandwidth reduction completion message has already been received is checked with reference to the resource supply completion node table 29 to check the serial number of the original transmission source node of the received all connection setting change completion message, Investigate the situation by comparing with the serial number in the setting change completion message.

In step 2103, if the message is already accepted with "YES", the serial number of the received all connection setting change completion message is determined by the serial number of the original connection setting change completion message received from the original transmission source node. If they are older or they match, it is determined that the reception has been completed, and the processing procedure ends.

When the serial number of the all connection setting change completion message received with “NO” in step 2103 is newer than the serial number of the original connection source node of the received all connection setting change completion message, and the message has not been received. In step 2104, the central control unit 24 updates the serial number of the target node in the resource supply completion node table 29.

FIG. 22 shows an example of the resource supply completion node table 29 in FIG. In FIG. 22, serial numbers “1208, 3305, 4720,...” Are recorded corresponding to the node identifiers “1, 2, 3,.

Further, in step 2105, the central control unit 2
Reference numeral 4 denotes a resource management table 28 among connections managed by using the resource management table 28, among connections in which the source node of the received all connection setting change completion message is the originating node. If the serial number is older than the received serial number of the complete connection setting change message, the active bandwidth is changed to the reduced bandwidth when the bandwidth was reduced, and the current serial number is changed to the serial number of the received complete connection setting change message. To change the settings.

Finally, in step 2106, the central control unit 24 sends an all connection setting change completion message to all adjacent nodes, and ends the processing procedure. Step 2106
The node identifier and the serial number of the all connection setting change completion message transmitted in step (1) are the same as the received message.

Next, with reference to FIG. 23 and FIGS. 15 to 18, a processing procedure in the case where the IF unit 1 detects the occurrence of a failure in a connection whose end point is the own node in the second embodiment. Will be described.

First, in step 2301, the IF unit 1 detects the occurrence of a failure in a connection connected to its own node, and sends a notification of the occurrence of a connection failure to the central control unit 24. As shown in FIG. 16, the connection failure notification is
It has the identifier of the failed connection as additional information. The connection identifier may be unique within this node.

Next, at step 2302, the central control unit 2
No. 4 refers to the resource management table 28 to check whether or not the own node is the originating node of the target connection. If “NO” in the step 2302 and the node is not the originating node, the processing procedure ends.

On the other hand, if “YES” is determined in step 2302 and the target connection is the originating node, in step 2303 the central control unit 24 records the number of times that the detour recovery processing of this connection has been attempted as a connection identifier. Band change counter N which is a counter
Is set to “1”.

Next, in step 2304, the central control unit 2
In step 4, the connection acceptability judging unit 25 refers to the resource management table 28 to search for a physical link next to the alternative connection, and checks whether the physical link can be accommodated in the next physical link even if the bandwidth of the alternative connection is increased. . It is assumed that the bandwidth of the alternative connection after the switching is specified in the resource management table 28 as the bandwidth at the time of the failure of the failed connection.

If “YES” in the step 2305 and the connection acceptability judging section 25 judges that the band can be increased, the central control section 24 controls the resource control section 2 in the step 2306 to control the fault. The connection is switched to the alternative connection and the bandwidth of the alternative connection is increased.
Change the settings related to the alternative connection.

Next, at step 2307, the central control unit 2
4 sends a connection switching request message to the next transit node of the alternative connection, and ends the processing procedure. In the connection switching request message, the connection identifier of the alternative connection is set as the connection identifier, the bandwidth increase width and the serial number are set as the bandwidth change width.

In the step 2305, when the connection acceptability judging section 25 judges that the increase of the bandwidth of the alternative route is impossible, the central control section 24 increments the value of the band change counter N which is the connection identifier in the step 2308. After that, in step 2309,
It is determined whether or not the value of the band change counter N has not reached the prescribed number.

[0200] If "NO" in the step 2309, the specified number of times has been reached, the processing ends as it is. In this case, the detour has failed.

On the other hand, if “YES” in the step 2309 and the specified number has not been reached, the central control unit 24 sets the time Tre to the timer 26 in a step 2310 and ends the processing procedure. Due to a timeout due to the time Tre of the timer 26, the own central control unit 24 performs the re-execution when receiving the connection switching request notification.

Next, a processing procedure when the central control unit 24 receives a connection switching request notification from the timer 26 in the second embodiment will be described with reference to FIG. 24 and FIGS.

First, in step 2401, the central control unit 24 receives a connection switching request notification from the timer 26. As shown in FIG. 16, the connection switching request notification has an identifier of the failed connection as additional information.

Next, at step 2402, the central control unit 2
In step 4, the connection acceptability judging unit 25 refers to the resource management table 28 to search for a physical link next to the alternative connection, and checks whether the physical link can be accommodated in the next physical link even if the bandwidth of the alternative connection is increased. .

If “YES” is determined in step 2403 by the connection acceptability judging section 25 to determine that the band can be increased, the central control section 2 determines in step 2404.
In step 4, the bandwidth management means switches the failed connection to the alternative connection and increases the bandwidth of the alternative connection, and the network control means changes the setting relating to the alternative connection in the resource management table 28.

Next, at step 2405, the central control unit 2
No. 4 sends a connection switching request message to the next transit node of the alternative connection and ends the processing procedure. In the connection switching request message, the connection identifier of the alternative connection as the connection identifier,
As the bandwidth change width, an increase width of the bandwidth and a common serial number with all connection setting change completion messages managed on a node basis are set.

If the connection acceptability judging unit 25 judges in step 2403 that the bandwidth increase of the alternative route is not possible due to “NO”, in step 2406, the central control unit 24 sets the band change counter N as the connection identifier. After the increment, in step 2407, it is determined whether or not the band change counter N has not reached a prescribed number. If “NO” in the step 2407, the specified number of times has been reached, the processing procedure ends. In this case, the detour has failed.

On the other hand, if “YES” in step 2407 and the specified number of times has not been reached, in step 2408 the central control unit 24 sets the time Trp in the timer 26 and ends the processing procedure. The central control unit 24 performs the re-execution when the connection switching request notification is received from the timer 26 due to the timeout due to the time Trp of the timer 26. It is assumed that the additional information of the connection switching request notification to be received is the same as the connection switching request notification received when this processing procedure is activated.

[0209] Except for the initialization of the band change counter N, this processing procedure is almost the same as the processing procedure in the case where the IF unit 1 detects a failure in a connection whose end point is its own node. That is, from step 2403 to step 24
Up to 08, it is completely the same as 2305 to procedure 2310 in FIG.

Next, a processing procedure when the IF unit 1 receives a connection switching request message from another node in the second embodiment will be described with reference to FIG. 25 and FIGS. 15 to 18.

First, in step 2501, the IF unit 1 receives a connection switching request message sent from another node. As shown in FIG. 16, the connection switching request message has, as additional information, an identifier of a connection whose bandwidth is to be changed, a bandwidth change width, and a serial number.

Next, at step 2502, the central control unit 2
Reference numeral 4 refers to the serial number of the connection that has received the bandwidth change request in the resource management table 28 and checks whether the connection that has received the bandwidth change request has the latest connection switching request message. If the message received in step 2503 is "NO", the processing procedure ends.

If the message received with "YES" in step 2503 is the latest, central control unit 24 determines in step 2504 whether or not the own node is the destination node for the connection for which the bandwidth change request has been received. .

If the connection for which the bandwidth change request has been received is the destination node in step 2504,
In step 2505, the central control unit 24 changes the bandwidth of the connection for which the bandwidth change request has been received from the resource management table 28 to make it the same as the connection switching request message for which the serial number has been received, and the connection for which the bandwidth change request has been received. Of the connection is switched by the resource control unit 2 on the basis of this information, the setting of the band and serial number of the target connection is changed in the resource management table 28, and the processing is performed. finish.

On the other hand, in step 2504, "NO"
In step 2506, if the connection that has received the bandwidth change request is not the destination node,
Investigates whether or not the bandwidth of the connection specified by the connection acceptability determination unit 25 can be accommodated in the next physical link when the bandwidth is changed by the specified width.

In step 2507, if “YES” can be accommodated in the next transit physical link, in step 2508 the central control unit 24 changes the bandwidth of the target connection by the resource control unit 2 and checks the target connection in the resource management table 28. Change the settings for the bandwidth and serial number.

Next, in step 2509, the central control unit 24
Sends a connection switching request message to the next via node following the target connection, and ends the processing procedure. At this time, the additional information stored in the connection switching request message has the same contents as the received connection switching request message when the connection identifier is unique within the communication network. When the connection identifier changes for each node, a converted numerical value is set for the connection identifier. The bandwidth change width and the serial number do not change.

When it is determined in step 2507 that the physical link cannot be accommodated in the next transit physical link in step 2507, step 25
At 10, the central control unit 24 sends the connection switching request message
A connection switching rejection message is transmitted, and the processing procedure ends. The connection identifier and the serial number set in the connection switching rejection message are the same as the received connection switching request message. The bandwidth change width is set by reversing the sign of the received connection switching request message.

Next, a process when the IF unit 1 receives a connection switching rejection message from another node will be described with reference to FIG. 26 and FIGS. 15 to 18.

First, in step 2601, the IF unit 1 receives a connection switching rejection message sent from another node and transfers it to the central control unit 24. As shown in FIG. 16, the connection switching rejection message includes, as additional information, the identifier of the target failed connection, the bandwidth for restoring the bandwidth, and the order of the bandwidth change processing at the source node of the connection. With the serial number shown.

In step 2602, the central control unit 24, having received the connection switching rejection message, refers to the serial number of the connection to be currently processed in the resource management table 28 and determines the received connection switching with respect to the target connection. Check if the rejection message is up to date. “NO” in procedure 2603
If the message is not the latest message, the processing procedure ends.

If “YES” is determined in step 2603 and the latest message is received, central control unit 2 is determined in step 2604.
In step 4, the bandwidth of the target connection is restored by the resource control unit 2 and the setting of the corresponding part is changed in the resource management table 28.

Next, at step 2605, the central control unit 2
No. 4 determines whether the own node is the destination node for the target connection. In step 2605, "N
If the target connection is not the destination node in O ", in step 2606 the central control unit 24 refers to the resource management table 28 to check the connection switching request source node, and The connection switching rejection message is sent to the via node, and the processing procedure is terminated.The bandwidth change width and serial number of the connection switching rejection message are the same as the received connection switching rejection message.The connection identifier is unique within the communication network. In that case, it is the same as the received message, and if it changes for each node, it changes accordingly.

On the other hand, if “YES” in step 2605 and the target connection is the destination node, in step 2607 the central control unit 24 refers to the resource management table 28 to identify and disconnect the connection before switching. After the return, the setting of the corresponding part is changed in the resource management table 28 based on the contents of the return.

Next, at step 2608, the central control unit 2
No. 4 counts the number of times of connection switching and increments a band change counter N serving as a connection identifier, and checks in step 2609 whether the band change counter N has not reached a specified number. If the specified number of times has been reached in “NO” in step 2609, the processing procedure ends. In this situation, connection switching failure is determined.

On the other hand, if “YES” in the step 2609 and the specified number is not reached, the central control unit 24 sets the time Trp in the timer 26 in the step 2610 and ends the processing procedure. When the time Trp of the timer 26 times out, the central control unit 24 receives a connection switching request notification from the timer 26 and starts a retry.

Next, with reference to FIG. 27 and FIGS. 15 to 18, a description will be given of the processing when the central control unit 24 receives a connection bandwidth increase request notification of a connection whose own node is an end point.

There is no particular limitation on the mechanism or reason for issuing the connection bandwidth increase request notification. Other than the reason that the bandwidth was reduced when the link failure notification or the resource supply request message was received, for example, the network management system adjusts the bandwidth in response to the traffic increase, changes in response to a request from a customer, And so on.

First, in step 2701, the resource control unit 2
Receives the connection bandwidth increase request notification, transfers the notification to the central control unit 24. As shown in FIG. 16, the connection bandwidth increase request notification includes, as additional information, the identifier of the connection for which the bandwidth is to be increased and the bandwidth after the bandwidth increase.

Next, in step 2702, the central control unit 24
Checks whether a valid resource supply flag is set when the connection bandwidth increase request notification is received. Step 27
If the valid resource supply flag is set in “03” in “03”, the processing procedure ends as it is.

On the other hand, if “NO” in the step 2703 and the effective resource supply flag is not set, the procedure 27
In 04, the central control unit 24 refers to the resource management table 28 by the connection acceptability judging unit 25 to check the next passing physical link of the connection, and determines whether it can be accommodated in the next passing physical link even if the connection bandwidth is increased. investigate.

If the connection acceptability determining unit 25 determines in step 2705 that the bandwidth cannot be increased by the connection acceptability determining unit 25, the processing procedure ends. On the other hand, Procedure 2
If “YES” in 705 and the connection acceptability determination unit 304 determines that the bandwidth can be increased, in step 2706 the central control unit 24 increases the bandwidth of the connection by the resource control unit 2 and The setting of the connection bandwidth and the serial number is changed in the resource management table 28.

Then, in step 2707, the central control unit 2
No. 4 sends a connection setting change request message to the next transit node of the target connection, and ends the processing procedure. In the connection setting change request message, the identifier of the target connection, the desired bandwidth change width, and the serial number indicating the order of the bandwidth change at the source node are set.

Next, the processing when the IF unit 1 receives a connection setting change request message from another node will be described with reference to FIG. 28 and FIGS. 15 to 18.

First, in step 2801, the IF unit 1 receives a connection setting change request message sent from another node and transfers it to the central control unit 24. The connection setting change request message includes, as shown in FIG.
With bandwidth change range and serial number,
This is a band change request.

At step 2802, the central control unit 24
The serial number of the connection that has received the bandwidth change request is searched in the resource management table 28 to check whether the connection setting change request message received for the connection that has received the bandwidth change request is the latest message. If the message received as “NO” in step 2803 is not the latest message, the processing procedure ends.

On the other hand, if the message received in step 2803 with “YES” is the latest message, step 280
In 4, the central control unit 24 determines whether or not the own node is the destination node for the connection for which the bandwidth change request has been received.

[0238] If "YES" in the step 2804 and the own node is the destination node, the central control unit 24 changes the setting of the bandwidth and the serial number of the target connection in the resource management table 28, and ends the processing procedure. I do.

On the other hand, if “NO” in step 2804 and the own node is not the destination node, central control unit 24 determines in step 2805 that connection acceptability judging unit 304 sets the connection designated as the post-switch connection. If the bandwidth is changed by the specified width,
Investigate whether it can be accommodated.

[0240] If "YES" is determined in step 2806 and the next pass physical link can be accommodated, the central control unit 24 changes the bandwidth of the target connection by the resource control unit 2 in step 2807, and The setting of the bandwidth and the serial number is changed in the resource management table 28.

Next, at step 2808, the central control unit 2
No. 4 sends a connection setting change request message to the next transit node of the target connection, and ends the processing procedure. When the connection identifier of the connection setting change request message is unique in the communication network,
When the identifier is the same as the received connection setting change request message and is unique only for each node and the identifier is converted each time the node passes, the value after the conversion is used. The bandwidth change width has the same value as the received connection setting change request message.

If it is determined in step 2806 that the physical link cannot be accommodated in the next transit physical link, in step 2809 the central control unit 24 sends the connection setting change request message to the change requesting node. A connection setting change rejection message is sent to the previous pass-through node toward, and the processing procedure ends. The connection identifier of the connection setting change rejection message is the same as the received connection setting change request message, and the bandwidth change width is set to a value obtained by reversing the sign of the connection setting change request message.

Next, a process in the case where the IF unit 1 receives a connection setting change rejection message from another node will be described with reference to FIG. 29 and FIGS.

First, in step 2901, the IF unit 1 receives a connection setting change rejection message sent from another node. As shown in FIG. 16, the connection setting change rejection message is a band change request having, as additional information, the identifier of the target connection, the band change width for restoring the bandwidth, and the serial number.

In step 2902, the central control unit 24 that has received the connection setting change rejection message refers to the serial number of the connection that has received the band change request in the resource management table 28 and receives the connection request that has received the band change request. Check whether the connection setting change rejection message is the latest one.

[0246] If "NO" in step 2903, and the received message is not the latest one, the processing procedure ends. On the other hand, if “YES” in the step 2903 and the received message is the latest, in the step 2904, the central control unit 24 returns the band of the target connection by the resource control unit 2 to the band of the target connection and the serial number. Is changed in the resource management table 28.

Then, in step 2905, the central control unit 24
Determines whether the own node is the destination node for the target connection. In step 2905, “YE
If the target connection is the destination node in S ", the processing procedure ends.

On the other hand, in the above procedure 2905, “YE
If the target connection is not the destination node in S ", in step 2906 the central control unit 302 sends a connection setting change rejection message to the change requesting node to the previous transit node of the connection, The bandwidth change width of the connection setting change rejection message to be transmitted is the same as the received connection setting change rejection message.
As in the case of receiving the connection setting change request message, if the message is unique within the communication network, it is the same as the received connection setting change rejection message, and if it changes for each node, it is changed accordingly.

The above is the description of the second embodiment.

In the second embodiment, the bandwidth reduction in the bandwidth reduction processing of the connection upon receiving the notification of the occurrence of the link failure is set in the resource management table before the failure occurs. Although the bandwidth is determined according to the bandwidth value at the time, the bandwidth can be determined by other methods.

For example, if the result of the bandwidth reduction for each connection is added as additional information to the all connection setting change completion message, it is possible to determine the failure status and the traffic status of each connection at that time. .

In one form of the resource management table shown in FIG. 30, the bandwidth reduction result for each connection is recorded in the bandwidth area instead of the three bandwidths in FIG. In addition, FIG. 31 shows the message type, the node identifier, the bandwidth change width (for each connection identifier), and the serial number as information held by the all connection setting change completion message.

Further, in the above embodiment, the bandwidth of the detour connection is fixed as the band value to be taken when the faulty connection has a fault. It is also possible to retry with a smaller bandwidth value each time it turns out to have failed. In this case, there is a high probability that the detour connection setting will succeed even if the time Trp, which is the time interval until the detour connection setting is retried, is “0”.

As other points to note, the same points as in the first embodiment are listed.

In the second embodiment, it is possible to use various methods that have been conventionally developed as connection setting methods for setting a detour connection, as in the first embodiment.

Also, the setting change for accommodating more connections in the physical link has been performed by reducing the bandwidth of the already set connection.
Other than this, there are methods such as quality control and release of some connections, which are the same as in the first embodiment.

Further, in this embodiment, the band of the failed connection is left as it is, and the band is not reused. However, this band is reusable as in the first embodiment.

Next, a third embodiment of the present invention will be described with reference to FIGS.

First, FIG. 32 shows the configuration of a node that executes failure recovery in the third embodiment. This node includes an IF unit 1, a resource control unit 2, a switch unit 3, a central control unit 34, a connection acceptability determination unit 5, a timer 36,
It comprises a resource supply flag area 27 and a resource management table 8. The connection acceptability judging section 5 and the resource management table 8 are connected.
4 is connected to each component and controls the main function of the node.

The nodes are connected by a physical link 11, and a connection 12 is accommodated in the physical link 11. Specific examples of the connection 12 include a path and a line, a virtual path and a virtual channel.

The node shown in FIG. 32 has almost the same configuration as the node shown in FIG. 15 in the second embodiment. The difference from FIG. 15 is that the material supply complete node table 29 (FIG. 15) Has been deleted. 1 in the first embodiment is that a resource supply flag area 27 is provided instead of the resource supply event management table 7 (FIG. 1). Components having the same functions as those of the nodes shown in FIGS. 1 and 15 are denoted by the same reference numerals, and description thereof is omitted.

The resource supply flag set in the resource supply flag area 27 is a flag indicating that the resource supply processing is being performed as in the case of the node shown in FIG.

Also in the third embodiment, the setting change of the connection so that more connections can be accommodated in the physical link is performed by reducing the bandwidth of the already set connection.

Next, an embodiment for executing the congestion recovery method based on the present invention in the configuration shown in FIG. 32 will be described with reference to FIGS.

First, FIG. 33 shows various notifications used for information transmission within the own node, which are used in this embodiment.
And information of various messages used for information exchange with other nodes.

In FIG. 33, there are a connection congestion occurrence notification, a congestion connection bandwidth increase request notification, and a resource supply flag clear notification as notifications. The connection congestion occurrence notification includes a message type, a connection identifier, and a bandwidth change width. The bandwidth increase request notification includes a message type and a connection identifier, and the resource supply flag clear notification includes a message type as additional information.

[0267] The messages include a resource supply request message, a connection setting change request message, and a connection setting change rejection message. The resource supply request message has a message type, a connection setting change request message, and a connection setting change rejection message. Includes a message type, a connection identifier, and a bandwidth change width as additional information.

Next, the congestion recovery of the connection in the above embodiment will be described with reference to FIGS.

First, in step 3401, the IF unit 1 detects the occurrence of congestion of the connection connected to its own node, and sends a connection congestion occurrence notification to the central control unit.

Next, in step 3402, the central control unit 34, having received the notification of the occurrence of the connection congestion, sets a 1-bit resource supply flag in the resource supply flag area 27. It does not have an expiration date or the like as in the resource supply flag in the second embodiment.

Next, in step 3403, the central control unit 3
4 sends a resource supply request message to the adjacent node. It is assumed that the resource supply request message has no additional information as shown in FIG.

Next, in step 3404, the central control unit 3
No. 4 uses the resource management table 28 to check whether or not there is a connection whose own node is the originating node. If “YES” is determined in step 3405, and there are connections that are source nodes, a process of reducing the bandwidth of those connections is performed in steps 3406 to 3409.

[0273] On the other hand, if "NO" in step 3405 and there is no connection that is the source node, the processing from step 3406 to step 3409 is skipped. The format of the resource management table 8 is the same as that of the first embodiment, that is, the format shown in FIG.

In the process of reducing the bandwidth of the connection, first, in step 3406, the central control unit 34 reduces the bandwidth of the target connection by the resource control unit 2 and uses the bandwidth by the network control means in the resource management table 8. Set to. The width of the decrease in the band may be determined in advance, or may be determined according to the connection usage status or the like at the time of this setting. If it is determined that it is not desirable to reduce the bandwidth of the connection when making the setting, it is not always necessary to reduce the bandwidth.

Next, in step 3407, the central control unit 3
No. 4 sends a connection setting change request message to the next transit node of the relevant connection. As shown in FIG. 33, the connection setting change request message has, as information, the identifier of the target connection whose bandwidth is to be changed and the change width. This change width may be the difference between before and after the change, or the bandwidth itself after the change.

Next, at step 3408, the central control unit 3
No. 4 checks whether the bandwidth reduction processing of all connections originating from the own node has been completed or not.
If the bandwidth reduction processing for all connections has not been completed at O ", the procedure returns to the above-mentioned step 3406 and repeats the procedure.

On the other hand, if “YES” in the step 3409 and the bandwidth reduction processing of all the connections is completed, the central control unit 34 clears the resource supply flag when a predetermined time has passed in a step 3410, 36
Is set to the time Tgr. When this time Tgr has elapsed, the timer 3
When the time-out occurs in 6, the central control unit 302 receives the resource supply flag clear notification from the timer 36 and clears the resource supply flag in the resource supply flag area 27.

Following the timer setting in step 3410, in step 3411, the central control unit 34 checks the position as to whether or not the own node is the originating node of the congested connection.
If “NO” in the step 3412 and it is not the calling side node, the processing procedure ends.

[0279] On the other hand, if "YES" in the step 3412 and the originating node is selected, the central control unit 34 sets the time Tsh in the timer 36 and ends the processing procedure in a step 3413. The central control unit 34 receives a congestion connection band increase request notification from the timer 36 due to the timeout of the time Tsh. The congestion connection band increase request notification has an identifier of the congested connection and a band increase width of the connection as additional information.

Next, a process in the case where the IF unit 1 receives a resource supply request message from another node will be described with reference to FIG. 35 and FIG. 32 and FIG.

First, in step 3501, the IF unit 1 receives a resource supply request message sent from another node. The resource supply request message has no additional information as shown in FIG.

Next, at step 3502, the central control unit 3
4 checks whether a valid resource supply flag is set in the resource supply flag area 27. If the flag is set to "YES" in step 3503, the processing procedure ends.

On the other hand, if the flag is not set at “NO” in step 3503, central control unit sets the resource supply flag in resource supply flag area 27 in step 3504, and sets all adjacent nodes in step 3505. The resource delivery request message is sent.

Next, at step 3506, the central control unit 3
4 checks whether there is a connection in which the own node is the originating node using the resource management table 8.
If “YES” is determined in step 3507 and the connection that is the originating node exists, steps 3508 to 35
Up to 11, processing for reducing the bandwidth of those connections is performed. On the other hand, if “NO” in the step 3507 and there is no connection that is the originating node, the processing steps from the steps 3508 to 3511 are skipped.

In the process of reducing the bandwidth of the connection, first, in step 3508, the central control unit 34
The resource control unit 2 reduces the bandwidth of the target connection, and the network control unit changes the setting of the resource management table 8 based on the bandwidth. The width of decrease of the bandwidth may be determined in advance, or at that time, the resource supply event management table, the use state of the connection, and the like may be referred to and determined accordingly.

Next, at step 3509, the central control unit 3
No. 4 sends a connection setting change request message to the next transit node of the target connection. As shown in FIG. 33, the connection setting change request message has, as information, the identifier of the target connection whose bandwidth is to be changed and the change width.

Next, in step 3510, the central control unit 34 checks whether or not the bandwidth reduction processing for all connections originating from the own node has been completed.
If the bandwidth reduction processing for all connections is not completed in O ", the procedure returns to step 3508 and repeats the procedure.

When the bandwidth reduction processing for all connections is completed in “NO” in step 3511, or when there is no connection that is the originating node in “NO” in step 3507, the central control unit 34 Tgr is set and the processing procedure ends. When a timeout occurs due to the time Tgr, the central control unit 34 receives a resource supply flag clear notification from the timer 36 and clears the resource supply flag in the resource supply flag area 27. The resource supply flag clear notification has no additional information.

Next, a description will be given of the processing when the central control unit 34 receives the congestion connection band increase request notification from the timer 36 with reference to FIGS.

First, in step 3601, the central control unit 34 receives a congestion connection band increase request notification from the timer 36. As shown in FIG. 33, the congestion connection band increase request notification has, as additional information, the identifier of the connection for which the band is to be increased and the band increase width.

Next, at step 3602, the central control unit 3
In step 4, the connection acceptability judging unit 5 refers to the resource management table 8 to check the physical link next to the connection, and checks whether the physical link can be accommodated in the physical link next even if the bandwidth of the connection is increased. If "NO" in the step 3603 and the connection acceptability judging unit 5 judges that the bandwidth cannot be increased, the processing procedure ends.

On the other hand, if “YES” in the step 3603 and the connection acceptability judging section 5 judges that the bandwidth can be increased, the central control section 34 in the step 3604 makes the resource control section 2 Is changed, and the setting of the corresponding portion relating to the target connection is changed in the resource management table 8.

Then, in step 3605, the central control unit 3
No. 4 sends a connection setting change request message to the next transit node of the target connection, and ends the processing procedure. In the connection setting change request message, the identifier of the target connection and the band change width are set.

Next, a process when the central control unit 34 receives a resource supply flag clear notification from the timer 36 will be described with reference to FIG. 37 and FIGS. 32 and 33.

First, in step 3701, the central control unit 34 receives a resource supply flag clear notification from the timer 36.

In step 3702, the central control unit 34, having received the notification of clearing the resource supply flag, clears the resource supply flag in the resource supply flag area 27, and ends the processing procedure.

The processing procedure when a connection setting change request message and a connection setting change rejection message are received is the same as the processing procedure described in the first embodiment with reference to FIGS. 7 and 8. Is omitted.

The above is the description of the third embodiment.

In the third embodiment, the setting change so that more connections can be accommodated in the physical link is performed by reducing the bandwidth of the already set connection. However, in the first embodiment, Similarly to the above, other methods such as quality control or release of some connections can be applied.

In the above description, the central control unit and the timer have been illustrated and described as components of different functions in each of the first to third embodiments, but all functions can be satisfied by one. . As described above, the above description has been made with reference to the drawings. However, it is possible to freely arrange functions by separating and merging the functions, change the order of the procedures before and after, and change the parallel operation as long as the above functions are satisfied. Is not limited.

[0301]

As described above, the following effects can be obtained by the present invention.

The first effect is that control can be speeded up.

[0303] The reason is that it is possible to perform distributed control of a communication network for recovering from a failure or congestion by using both a spare resource and a resource that is not currently required at a high level. This is because it is not necessary to exchange a large amount of information between the center and each node.

A second effect is that it is possible to easily cope with a case where the communication network is enlarged.

The reason is that control is performed in a distributed manner by each node and a centralized center is not required for the above-mentioned reason, so that a situation in which the load on the center increases and the control becomes difficult does not occur.

A third effect is that the entire control is not stopped by a single failure.

The reason is that, similarly to the above-mentioned reason, there is no centralized center and the control is distributed.

[0308] A fourth effect is that the mounting of the components is facilitated.

The reason is that there are many common algorithms for the process of detouring due to a failure and the process of recovery from congestion, and it is possible to handle both in a unified manner.

[Brief description of the drawings]

FIG. 1 is a functional block configuration diagram of a node according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing an example of a notification and a message transmitted and received in FIG.

FIG. 3 is a configuration diagram illustrating an example of a resource supply event management table in FIG. 1;

FIG. 4 is a configuration diagram showing one embodiment of a resource management table in FIG. 1;

FIG. 5 is a flowchart showing an example of a recovery processing procedure when a link failure occurs in FIG. 1;

FIG. 6 is a flowchart illustrating an example of a resource supply request procedure of the failure recovery processing in FIG. 1;

FIG. 7 is a flowchart illustrating an example of a connection setting change request procedure of the failure recovery processing in FIG. 1;

FIG. 8 is a flowchart illustrating an example of a connection setting change rejection procedure of the failure recovery processing in FIG. 1;

FIG. 9 is a flowchart illustrating an example of a connection failure detection procedure in the failure recovery processing in FIG. 1;

FIG. 10 is a flowchart illustrating an example of a connection switching request notification procedure of the failure recovery processing in FIG. 1;

FIG. 11 is a flowchart illustrating an example of a connection switching request procedure of the failure recovery processing in FIG. 1;

FIG. 12 is a flowchart showing an example of a connection setting change rejection procedure of the failure recovery processing in FIG. 1;

FIG. 13 is a flowchart showing an example of a procedure for notifying all connection resource acquisition permission of the failure recovery processing in FIG. 1;

FIG. 14 is a configuration diagram showing one embodiment of a resource management table different from that of FIG. 4;

FIG. 15 is a functional block configuration diagram of a node according to the second embodiment of the present invention.

FIG. 16 is an explanatory diagram showing an example of a transmission / reception notification and a message in FIG.

17 is a configuration diagram illustrating an example of a resource supply flag in FIG.

18 is a configuration diagram illustrating an example of a resource management table in FIG.

FIG. 19 is a flowchart showing an example of a recovery processing procedure when a link failure occurs in FIG.

FIG. 20 is a flowchart illustrating an example of a resource delivery request procedure of the failure recovery processing in FIG.

FIG. 21 is a flowchart illustrating an example of a procedure for completing all connection setting changes in the failure recovery processing in FIG. 15;

FIG. 22 is a configuration diagram showing an example of a resource supply completion node table in FIG.

FIG. 23 is a flowchart illustrating an example of a recovery processing procedure when a connection failure occurs in FIG. 15;

24 is a flowchart showing an example of a connection switching request notification procedure of the failure recovery processing in FIG.

FIG. 25 is a flowchart illustrating an example of a connection switching request procedure of the failure recovery processing in FIG. 15;

26 is a flowchart illustrating an example of a connection switching rejection procedure of the failure recovery processing in FIG.

FIG. 27 is a flowchart illustrating an example of a connection bandwidth increase request notification procedure of the failure recovery processing in FIG. 15;

FIG. 28 is a flowchart illustrating an example of a connection setting change request procedure of the failure recovery processing in FIG. 15;

FIG. 29 is a flowchart illustrating an example of a connection setting change rejection procedure of the failure recovery processing in FIG. 15;

FIG. 30 is a configuration diagram showing one embodiment of a resource management table different from that of FIG. 18;

FIG. 31 is an explanatory diagram showing an example of an additional message in FIG. 16;

FIG. 32 is a functional block configuration diagram of a node according to the third embodiment of the present invention.

FIG. 33 is an explanatory diagram showing an example of a transmission / reception notification and a message in FIG. 32;

FIG. 34 is a flowchart illustrating an example of a recovery processing procedure when connection congestion occurs in FIG. 32;

FIG. 35 is a flowchart illustrating an example of a resource delivery request procedure of the congestion recovery processing in FIG. 32;

36 is a flowchart showing an example of a congestion connection band increase request notification procedure of the congestion recovery processing in FIG. 32.

FIG. 37 is a flowchart showing an example of a resource delivery flag clear notification procedure of the congestion recovery processing in FIG. 32.

FIG. 38 is a conceptual diagram of a conventional fault recovery control method using centralized control using working resources.

[Explanation of symbols]

 1 IF (Interface) Unit 2 Resource Control Unit 3 Switch Unit 4, 24, 34 Central Control Unit 5, 25 Connection Acceptability Determination Unit 6, 26, 36 Timer 7 Resource Delivery Event Management Table 8, 28 Resource Management Table 11 Physical Link 12 Connection 27 Resource Delivery Flag Area 29 Resource Delivery Complete Node Table

Continuation of the front page (58) Fields investigated (Int. Cl. ⁷ , DB name) H04M 3/22 H04M 3/00 H04M 3/36

Claims (11)

(57) [Claims] 1. A connection type communication network in which a plurality of nodes are connected by physical links, and each physical link accommodates a connection forming a communication path via at least one physical link. In the failure recovery method when a failure occurs in each node, each node has a resource management table that manages the settings of the connection passing through its own node, and the node that detects the failure on the physical link connected to its own node A resource delivery request message containing the location information of the failure is sent to all the neighboring nodes. When the resource delivery request message received is the first message related to the failure, the resource delivery request message is sent to all the neighboring nodes. The resource sending request message is sent to the node. Of the connections from which the local node is the transmission source, for the connections for which no failure has been detected, change the connection settings so that more connections can be accommodated on the physical link, and the connection settings that include the changed settings A change request message is sent to the next transit node following the connection. If the node that has received the connection setting change request message is a relay node related to the connection, the node determines the connection based on the received connection setting change request message. The connection setting change request message is transferred to the next node by changing the setting of the resource management table regarding the connection. If the own node is the destination node for the connection, the connection is changed based on the connection setting change request message. Changes the setting of the resource management table regarding the connection, if the node is the transmission source of the connection, the node detects the connection in which the failure is detected through a path different from that before the occurrence of the failure by means of communication between the nodes. If the connection setting needs to be changed in order to use more network resources, the connection is established after a predetermined period of time has elapsed after receiving the resource supply request message. A failure recovery method in a connection-type communication network characterized in that a setting change is attempted. 2. A connection type communication network in which a plurality of nodes are connected by physical links, and each physical link accommodates a connection forming a communication path via at least one physical link. In the failure recovery method when a failure occurs, each node has a resource supply flag setting area indicating that the resource supply processing of the connection is being performed, and a resource for managing the setting of the connection passing through the own node. A management table, and the node that has detected the failure of the physical link connected to its own node sends a resource supply request message including the location information of the failure to all adjacent nodes when a valid resource supply flag is not set. And the node receiving the resource supply request message also sets the valid resource supply flag unset. The nodes the resource dispensing request message sent to all the neighboring nodes, and sends the resource dispensing request message, the
For connections for which the local node is the transmission source and for which no failure has been detected, change the connection settings so that more connections can be accommodated on the physical link, and use the connection that includes the changed settings. Sends a setting change request message to the next transit node following the connection.If the node that has received the connection setting change request message is a relay node related to the connection, based on the received connection setting change request message, The connection setting change request message is sent to the next node by changing the setting of the resource management table for the connection, and if the own node is the destination node for the connection, the connection setting change request message is sent based on the received connection setting change request message. Changes the setting of the resource management table relating to the connection, and when the node is the transmission source of the connection, the node distributes a different path to the connection where the failure is detected by communication between the nodes before the occurrence of the failure. If it is necessary to change the connection setting in order to use more network resources, a predetermined time after receiving the resource supply request message elapses and thereafter, A failure recovery method in a connection-type communication network characterized by attempting to change a connection setting. 3. A physical link in a connection-type communication network in which a plurality of nodes are connected by physical links, and each physical link accommodates a connection forming a communication path via at least one physical link. In the failure recovery method when a failure occurs, each node has a resource management table that manages the setting of the connection passing through the own node, and the node that detects the failure of the physical link connected to the own node The resource delivery request message including the location information of the failure is sent to all the adjacent nodes, and the node receiving the resource delivery request message sends the resource delivery request message to all the adjacent nodes if the received resource delivery request message is the first message related to the failure. Sends this resource offer request message to the node, and the node that sends this resource offer request message , Change the setting so that more connections can be accommodated by the physical link among all connections for which no failure has been detected among the connections from which the own node is the transmission source, and complete when the setting change for all connections is completed. The node sends all connection setting change completion messages for notifying the completion of the setting change to all nodes adjacent to its own node. If the source node is the first node, the node sends the received all connection setting change completion message to all adjacent nodes, and the node determines whether a failure has been detected in the source connection. Tries to distribute the connection differently than before the failure by connecting the nodes When the setting is changed, and the connection setting needs to be changed in order to use more network resources, the connection setting change is performed after a predetermined period of time has elapsed after receiving the resource supply request message. A failure recovery method in a connection-type communication network characterized by trying. 4. A connection type communication network in which a plurality of nodes are connected by physical links, and each physical link contains a connection forming a communication path via at least one physical link. In the failure recovery method when a failure occurs, each node has a resource supply flag setting area, and a resource management table for managing the setting of a connection passing through the own node, and the physical connection connected to the own node. When the valid resource supply flag is not set, the node detecting the occurrence of the link failure sends a resource supply request message to all adjacent nodes, and similarly, the node receiving the resource supply request message also supplies a valid resource supply message. If the flag is not set, the resource delivery request message is sent to all adjacent nodes, and this resource delivery request message The node that sent the message changes the settings so that more connections can be accommodated on the physical link for all connections for which no failure has been detected among connections originating from the own node. When the setting change is completed, all connection setting change completion messages for notifying the completion of the setting change are sent to all nodes adjacent to the own node. If the original transmission source node of the connection setting change completion message is the first node, send all the received connection setting change completion messages to all the adjacent nodes, and if the node is the source of the connection, For the connection in which the failure was detected, a different path between the nodes was used, which was different from the path before the failure. If it is necessary to change the setting of the connection in order to use the network resources more in a distributed manner, after a predetermined period of time has passed after receiving the resource supply request message, A failure recovery method in a connection-type communication network, characterized by attempting to change the connection setting. 5. The node according to claim 1, 2, 3, or 4, wherein the node attempts to set up a connection via another path when the connection to which the node is a transmission source has failed. A failure recovery method in a connection-type communication network, which is performed after a lapse of a predetermined time from the detection of the occurrence. 6. The method according to claim 1, wherein the node sets in advance an attempt to establish a connection by another route when a connection from which the node is a transmission source has failed. A failure recovery method in a connection-type communication network, wherein the method is repeated a predetermined number of times. 7. The resource management table according to claim 1, wherein the resource management table of the node includes a resource amount currently used by a connection passing through the node, a resource amount used in a normal state, and a resource amount used in a failure occurrence. It has an area for storing the amount of resources to be used, the information for identifying the previous node through which the connection passes, the information for identifying the next node, and the identifier of the alternative connection. A failure recovery method in a connection-type communication network, wherein the bandwidth of the resource management table is changed to a resource amount at the time of failure in the resource management table. 8. A connection-type communication network in which a plurality of nodes are connected by physical links, and each physical link contains a connection forming a communication path via at least one physical link. In the congestion recovery method when congestion occurs, each node has a resource management table that manages the settings of connections passing through the own node, and the node that detects the occurrence of congestion of the connection connected to its own node sends a resource delivery request. Sends a message to all adjacent nodes, and the node that received this resource offer request message forwards this resource offer request message to all adjacent nodes if the resource offer request message received for the congestion is the first message. Then, the node that sends this resource supply request message is For the connection with the setting, change the setting so that more connections can be accommodated on the physical link, and send a connection setting change request message for transmitting the changed setting to the next transit node of the connection. If the own node is the relay node for the connection, the node that received the request message changes the setting of the resource management table for that connection based on the connection setting change request message, and sends this connection setting change request message to the next node. If the own node is the destination node for the connection, the node changes the setting of the resource management table for the connection based on the connection setting change request message, and each of the nodes first detects the occurrence of congestion. Attempt to increase the bandwidth for the connection, and if it is necessary to change the connection setting to use more network resources, a predetermined period of time has elapsed after receiving the resource supply request message A congestion recovery method in a connection-oriented communication network, characterized in that a connection setting change is attempted later. 9. A connection type communication network in which a plurality of nodes are connected by physical links, and each physical link accommodates a connection forming a communication path via at least one physical link. In the congestion recovery method when congestion occurs, each node includes a resource supply flag setting area and a resource management table for managing the setting of a connection passing through the own node. The node that detects the occurrence of congestion sends a resource offer request message to all adjacent nodes when the valid resource offer flag is not set, and similarly, the node that receives this resource offer request message also has a valid resource offer flag. If is not set, send a resource delivery request message to all neighboring nodes, and The sending node changes the connection setting so that more connections can be accommodated in the physical link with respect to the connection from which the own node is the transmission source, and issues a connection setting change request for transmitting the changed setting content. The node that has received the connection setting change request message sends the message to the next transit node following the connection, and if the own node is a relay node for the connection, the node related to the connection based on the connection setting change request message The setting of the resource management table is changed, and this connection setting change request message is sent to the next node. If the own node is the destination node for the connection, the connection setting change request message is sent based on the connection setting change request message. Each node should attempt to increase the bandwidth for the connection that first detected the occurrence of congestion, and need to change the connection setting to use more network resources. In this case, a connection setting change is attempted after a predetermined period of time has elapsed after receiving the resource supply request message. 10. The node according to claim 8, wherein the node detects an occurrence of congestion related to the connection by attempting to increase the bandwidth of the connection performed when the connection from which the node is a transmission source is congested. A congestion recovery method in a connection-type communication network, wherein the method is performed after a certain time has passed since the start of the communication. 11. The node according to claim 8, wherein the node repeats, for a predetermined number of times, an attempt to set a bandwidth increase of a connection performed when the connection from which the node is a transmission source is congested. Congestion recovery method in connection type communication network.