JPH0213154A - Bypass route selection processing system - Google Patents

Abstract

PURPOSE:To allow each node to simply select a bypass route automatically when a fault takes place in a route used for communication by referencing a route list table in the communication with other node, selecting a route through which fault display information is usable and applying the communication. CONSTITUTION:A route list management section 13 applies the processing to set fault display information of route enable/disable use as to a relevant route on a route list table 14 with respect to a notice from a fault notice section 12. A communication processing section 15 references the route list table 14 at the communication with other node and a route through which fault display information is usable is selected and a communication processing is executed. Thus, when a fault takes place in the route used for the communication, each node is able to select the bypass route simply automatically.

Description

[Detailed Description of the Invention] [Summary] This invention relates to a detour route selection processing method when a route failure occurs in a communication system that performs data communication between nodes in a computer network according to a predetermined route. The purpose is to automatically and easily provide a means for each node to select a detour route.

Each node has a route list table that has a list of elements that make up the route to the destination node for each route to that destination node, and the failure occurrence and recovery of the elements that make up the route between it and other nodes. and a route list management unit that sets failure display information on route usability/unavailability for the corresponding route in the route list table based on the information from the failure notification unit. In preparation, when communicating with other nodes, the route list table is referred to, and a route whose failure display information is set to "ROU" is selected for communication.

[Industrial application field]

The present invention provides a 1lf communication that performs data i1m communication according to a predetermined route between nodes in a combination 1-tane node work.
The present invention relates to a processing method for selecting a detour route when a route failure occurs in a communications system.

[Conventional technology]

FIG. 5 shows an example of a computer network, and FIG. 6 shows an example of route management information according to a conventional method.

BACKGROUND OF THE INVENTION With the development of computer usage technology, computer networks have come to be used in various forms in a sophisticated manner.

FIG. 5(a) is an example of the computer network. In FIG. 5(A), 4OA to 40C represent host computers, 41A to 41E represent communication processing devices having a communication control function, and 42A to 42E represent terminal devices.

For example, between host computer 40A and host computer 40C,
48, the communication processing device 41B acts as a relay node.
There are various routes such as a route passing through the communication processing device 41D and a route passing through one communication processing device 41D. host computer 40
The same applies when communicating between A and the terminal device 42E.

For the purpose of explaining the present invention, various such materials are used.

An example of a simplified network in which each communication processing device or a combination of a communication processing device and a host computer is represented as a node, and lines connecting nodes are represented as links is the network shown in Figure 5 (b). be. In FIG. 5(b), A to D represent nodes having data transmission/reception and relay functions, and L1 to L5 represent links such as lines.

There are three types of communication routes from node A to node D: a route that passes through node C, a route that goes directly to node D, and a route that passes through node B. Each route is designated as route #1. We will call them route #2 and route #3.

Traditionally, management (definition) information regarding such routes is
Each node had distributed information as shown in FIGS. 6(a) to 6(d).

When making a call, the two calling nodes pass a call request in which the path number to be used and the destination node are set to the node defined in the route information held by the own node.

The relay notebook relays the call request based on the information set in 5 call requests and refers to the route information held by its own node.

As an example, a case where node A communicates with notebook D using route #3 will be explained based on the J management information shown in FIG.

Calling node A refers to its own route definition information,
5 (path P5. path P6) for route #3, and [destination node: node Di! Bus: Pass P5. Create a call request with bus P6J set. Next, from the path definition information, find the destination for the data on path P5 leading to node D, head to node B, link L
A call request is sent through 2.

Relay node B focused on the used path P5 among the 9 call requests, and based on the path definition information it had, headed to node D, sent 2 call requests through link L5, and relayed 1 call request. Record what happened.

When the called node D receives a call request from the calling node A, it creates a response to the request.

In response to a call, [destination node: node D, path used: path P5. Set information called bus P6J.

Next, from the path definition information shown in FIG. 6(d).

Path l leading to node A] Finds the destination to which the data on G is sent, heads to node B, and sends out a call response through link L5. When called node D returns two call responses, it considers that the route between nodes A and D has been determined.

Again, the relay node B focuses on the bus P6 used for the two-call response request, and sends the call response to the node A through the link L2 based on the path definition information it has. The relay node B considers that the route between the node and the node D is determined when one call response is relayed.

When calling node A receives one call response, it recognizes that the route has been determined.

When a failure occurs in a node or a link, the node that discovered the failure uses its own definition information.

The information that "the path number" from the self to "where" has become unusable is notified to the adjacent node. The node that received the notification uses that information and the definition information it owns,
It creates the sentiment that ``the bus number'' leading from me to ``where'' has become unusable, and similarly notifies other adjacent nodes. Furthermore, determine which route in your own definition information has become unusable.

For example, if a failure occurs in link L5 shown in FIG. 5(b), node B detects the failure in link L5, and then
From the definition information it owns, it notifies the adjacent node (here, node A) that "the bus P5 from itself to node D has become unusable."

Node A, which received the notification, uses the notification and definition information to
Recognize that route #3 is unusable.

Further, from the definition information, the node C is notified that "the path P5 from itself to the node D has become unusable." This notification is used at node C as well.

[Problem to be solved by the invention]

In the conventional route-related definition information management described above, a route is associated with a path, and one calling node, relay node, etc.
Since the method determines the destination and the link to be used from the information of the destination node and the bus to be used, the following problems arise.

(al Current route and alternative route in case of failure (
Bank up route) is difficult to define.

The algorithm/protocol for notifying other nodes of the occurrence of a fbl route failure is complex.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems and provide a means for each node to automatically and easily select a detour route when a failure occurs in a route used for two communications.

[Means to solve the problem]

FIG. 1 is a block diagram of the principle of the present invention.

In FIG. 1, 11 is a fault monitoring unit that monitors the occurrence of a detectable fault in its own node, 12 is a fault notification unit that processes fault notifications and recovery notifications from the fault monitoring unit 11 and other nodes, and 13 is used for communication. 14 is a route list table; 15 is a communication processing unit that performs a process of establishing a connection with a destination node; A-D are nodes; LL to L5 are links.

Each node A to D, for each route to each destination node,
It has a route list table 14 that stores a route list consisting of the elements constituting the route to the destination node, that is, the arrangement of links, relay nodes, etc.

When the failure notification unit 12 receives a failure notification of an element constituting a route or a failure recovery notification from the failure monitoring unit IT or another node, it performs a process of notifying other adjacent nodes of the failure occurrence or recovery information. I do. Also, to the route list management section 13.

Notify the information.

The route list management unit 13 performs a process of setting failure display information indicating whether the route can be used or not for the corresponding route in the route 1-list table 14 in response to the notification from the failure notification unit 12.

When communicating with other nodes, the communication processing unit 15 refers to the route list table 14, selects a route whose failure display information indicates that the route can be used, and executes communication processing.

[Effect]

In the present invention, a route list method is adopted for route definition in order to allow troops to define the current route/alternative route and determine whether the route can be used in the event of a failure. Route definition using the route list method means that each node holds all available routes starting from itself in the form of a list table.

For example, node A has definition information based on a route 1-list, such as the route list table 14 shown in FIG. 1, in order to manage the routes it uses. The route list contains information about the nodes and links that a specific route passes through.
Everything is specified.

When the communication processing unit 15 establishes a connection, the starting point node sets a route list of routes to be used in the request for establishing a connection. Each node refers to this route list and sets up a connection.

In addition, the route failure notification algorithm and block 1-
11-In order to simplify the route, in the present invention, instead of notifying each node whether the route can be used or not,
Each node is notified of the availability/unavailability of nodes/links that are constituent elements of the route.

Then, the determination of whether the route itself can be used is determined by the node/
Each node performs this individually in accordance with the failure display information set in the route list table 14 based on the link availability/unavailability information.

〔Example〕

FIG. 2 is a configuration example of a route list table according to an embodiment of the present invention, FIG. 3 is a detailed block diagram of a failure notification section according to an embodiment of the present invention, and FIG. 4 is a diagram according to an embodiment of the present invention. A detailed block diagram of a route list management unit and a communication processing unit is shown.

[Description of Route List I Table] Each node has one route list table 14 as shown in FIG. 2 in this embodiment. A pointer to the route list table. and the identifier (ID) of the destination node.

It stores a flag indicating whether or not a route can be used, the number of nodes/links composing the route, and other attached information, and information on a route list with a flag for indicating a failure for each element.

In the route list, a flag for indicating a failure is "0" when it is normal, and "l" is set for a node or link that is a failure. Note that a flag is set as a failure indication for each element in the route list, so I! 7a: < , each node has 4
1. An identifier for a given failure (referred to as a failure ID) may be set.

The route usability flag indicates a failure regarding the route, and when it is "l", it is unusable due to a failure.

“0” indicates usable. When a failure indication is set for any one or more of the nodes or links in the route list, the route usability flag for that route is set to "l" and the route is disabled.

Of course, the route list table 14 is not limited to the configuration shown in FIG. Other configurations may be used to hold information regarding the route list.

[Description of Fault Monitoring Unit] The fault monitoring unit 11 shown in FIG. 1 is a part that monitors the status of nodes/links adjacent to the current node. The part that sends and receives data, such as the communication processing unit 15, is
5 It is recognized that a failure has occurred when normal input/output (1/(J)) cannot be performed. If a failure occurs and communication is no longer possible, 2 that information + a is notified to the failure notification unit 12ζ1. On the other hand, even if something in a faulty state is restored to a normal state, the failure ifll knowledge unit 12 sends a recovery notification to that effect.
send to

[Description of Failure Notification Unit] The failure notification unit 12. It has a processing function to exchange information with other nodes. For example, it may be configured as shown in FIG.

The interface 20 includes the fault monitoring unit 1j and the route list! - Exchange information with the management section 13.

The other node I, . . . /0 processing unit 21 performs processing for setting up/terminating a communication connection with a fault notification unit in another node and communication processing using the connection.

The router 22 is. It determines how to process input information and executes various processes according to the input information, which will be described later.

The connection management unit 23 manages the presence or absence of a connection with a failure notification unit of another node.

The fault ID counters 2 and 4 are numbering counters for assigning a unique identifier (fault ID) to a fault discovered in the own node.

The storage device 25 is used here for two purposes. It is used to store information for associating failure IDs with failure locations, and to save information regarding control connections with other nodes.

The failure notification unit 12 performs the following initial settings in advance to select a route when a certain node attempts to start communication processing.

In the failure notification section 12, control is passed to the connection management section 23 at the time of initial setting. The connection management unit 23 sets up control connections with other nodes, and also prepares an area in the storage device 25 for storing information regarding the connections.

There are four types of secondary inputs to the fault notification unit 12 during normal operation.

1) Failure notification from failure monitoring unit 11 2) Recovery notification from failure monitoring unit 11 3) Failure notification from other nodes 4) Recovery notification from other nodes When these are input, the failure notification unit 12:

Each is processed as follows.

l) Processing for failure notifications from the failure monitoring unit.

■The router 22 has an ink facer 20 output.

The failure notification from the failure monitoring unit 11 is received.

(2) The router 22 refers to the failure ID counter 24 and determines the failure ID to be assigned to the failure for which the first notification has been received. Then, the location of the failure and the failure ID are combined and stored in the storage device 25.

(2) The router 22 notifies the route list management unit 13 via the interface 20 of the occurrence of a node/link failure.

(2) Next, the router 22 inquires of the connection management unit 23 about the nodes connected to the control connection.

■ For the node obtained by inquiry, other nodes 110
5 failure information is notified via the processing unit 21. Storage device 2
Failure 10 occurs in the storage area for the corresponding connection in 5.
Set.

2) Processing for recovery notification from the failure monitoring unit.

■ Router 22 is Inkfacer 20 Kyoyama.

A recovery notification is received from the failure monitoring unit 11.

(2) The router 22 refers to the notification 25 and obtains the failure ID assigned to the recovered failure.

(2) The router 22 notifies the route list management unit 13 via the interfacer 20 of node/link failure recovery.

(2) Next, the router 22 refers to the storage device 25.

hindrance! Find the node that notified the failure corresponding to D.

■ For the found node, other node I10 processing unit 2
Notify the recovery information of 3 failures via 1.

Clear the failure ID of the corresponding connection storage area in the storage device 25.

3) Processing for failure notifications from other nodes.

■ The router 22 has an output from the other node I10 processing unit 21,
Receives failure notifications from other nodes.

■ Router 22 is Inkfacer 20 Kyoyama.

The route list management unit 13 is notified of the occurrence of a failure in 9 nodes/links.

4) Processing for recovery notifications from other nodes.

■ The router 22 is the other node Xia 10 processing unit 21 Keishan,
Receives failure recovery notifications from other nodes.

(2) The router 22 notifies the route list management unit 13 via the interfacer 20 of node/link failure recovery.

[Description of Route List Management Unit 1 As shown in FIG. 4, the route list management unit 13 includes a failure setting processing unit 30 that is activated by a failure notification from the failure notification unit 12, and a failure reset processing unit 30 that is activated by a recovery notification. It has a processing section 31. The process is as follows.

l) Processing of the fault setting processing unit 30.

■ Search the route list in the route list table 14 to see if there is a route list that uses the node/link where the failure occurred.

- If a route list using the node/link at the point where the failure has occurred is found, failure display information is set for the corresponding node/link, and first, the route usability flag for that route is set to "l".

(2) The above process is repeated for all rules l- in the route list table 14.

2) Processing of the failure h/T removal processing unit 31.

■ In the route list in the route I list table 14,
Search to see if there is anything that uses the 113 notes/links for failure recovery.

■ Fault iu 11. If you find a route list that uses one node/link, 5 do not erase the fault display for the corresponding node/link.

■ If the fault display is cleared. Search for other fault indications in the route list, and go through three cycles of the node/link where the fault is occurring. If there is no node/link in which a failure has occurred, the route availability flag for that route is reset to "0".

② Repeat the above-mentioned process for all routes in the route list table 14.

For example, from the failure notification unit 12. When there is a notification that link L5 has entered the failure state, the route 1-IJ strike management unit 13 searches the entire route list in the route list table 14 to find link L5, and Set the link failure display to indicate that it is in a failure state. Additionally, the route is not usable. Set and display route availability flag.

Conversely, if there is a notification of failure recovery. Search the entire route list. Clear the fault display for the corresponding node/link.

Furthermore, if no other nodes/links on the route are in a failure state, the route availability flag is cleared.

[Description of J communication processing section] The communication processing section 15 has a route list table reference processing section 32 and a connection 6' existence processing section 33, as shown in FIG.

■) Processing of the route table reference processing unit 32.

The route list 1-list reference processing unit 32 is activated by the connection establishment processing unit 33 to refer to the route list 1-table 1.
Perform the search for 4 and find the destination node.

Search for route list 1- which is the same as that specified by the connection establishment processing unit 33 and whose route usability flag is "0°'". →) If a corresponding route list is found as a result of searching, the information is passed to the connection establishment processing section 33.

2) Processing of the connection establishment processing unit 33.

The connection establishment processing unit 33 performs connection establishment processing for logical connection with a communication destination.

To select the route used to establish a connection. The route table reference processing unit 32 is called.

Once the route list is obtained, the processing section 33 performs the following operations. For requesting a call to the other party's notebook.

Set the information for this root list 1-. Perform call processing.

As mentioned above, each node has a route list table 14 that holds route definition information based on a route list, and a failure notification unit 12 has a route list table 14 that holds route definition information based on a route list. The route list management unit 13 notifies each other of node/link failure occurrence or recovery.
For example, based on information from the failure notification unit 12. Set information on whether routes can be used or not. As a result, when a failure occurs, detour route I. is automatically selected.

Note that a failure in one root component is detected by multiple other nodes. Failure notifications may come from multiple nodes. In this case, the recovery process will be based on the initial failure recovery notification. You may clear the fault indication for that node/link. Wait until there are multiple recovery notifications that correspond to the failure notification.

Then, the failure display may be canceled.

In that case, for example, the identifier of the node that sent the failure notification and the failure ID may be stored in the storage device 25 as bare information. It is possible to manage correspondence between failure notifications and recovery notifications. The failure ID may also be entered in the route list.

In addition, the count control for the number of failure notifications can be
Correct the link failure and reduce the count with a recovery notification. It may be determined that the normal state has returned when the state becomes O.

〔Effect of the invention〕

As explained above. According to the present invention. Since each node has a route list corresponding to the route to the destination node,
Route selection becomes very easy.

Since the route list expresses routes in a format similar to the routes that humans think of, it becomes much easier to define the current route I.1 alternative route in Yuji's network generation, etc. Furthermore, by using nodes/links, which are constituent elements of a route, as the unit for notifying the occurrence/recovery of one failure, the algorithm and protocol for determining whether a route can be used is simplified. This makes it easier to keep track of the routes in use and available routes.

[Brief explanation of the drawing]

FIG. 1 shows the basic principle of the present invention. FIG. 2 shows an example of the configuration of a route list table according to an embodiment of the present invention. FIG. 3 is a detailed block diagram of a failure notification section according to an embodiment of the present invention. FIG. 4 is a detailed block diagram of an IJ storage management section and communication processing section according to an embodiment of the present invention. FIG. 5 is an example of a computer network. FIG. 6 shows an example of route management information according to the conventional method. In the figure, A-D are nodes and Ll-L5 are links. Reference numeral 11 represents a fault monitoring section, 12 a fault notification section, 13 a route list management section, and 14 a route list full 15 a communication processing section.

Claims (1)

[Claims] In a communication system that performs data communication between nodes in a computer network according to a predetermined route, each of the above-mentioned nodes (A, B, C,...) transmits data for each route to each destination node. A route list table (14) that has a route list based on the arrangement of elements that make up the route to the destination node, and a failure notification unit (12) that notifies the occurrence and recovery of failures in the elements that make up the route between other nodes. and a route list management unit (13) that sets failure display information on whether the route can be used or not for the corresponding route in the route list table (14) based on the information from the failure notification unit (12). A detour route characterized in that, when communicating with another node, the route list table (14) is referred to and a route whose failure display information indicates that the route can be used is selected for communication. Selection processing method.