JPH07160601A - Communication equipment replacement processing system - Google Patents

Abstract

PURPOSE:To provide a communication equipment replacement processing system which can dynamically grasp the states of communication equipments included in the decentraized processing system and can perform the replacement processing without affecting the system despite occurrence of a fault by multiplexing the communication groups of the decentralized processing system and using a device control table which controls the information on the communication equipments attendant on the nodes that construct the decentralized processing system. CONSTITUTION:This processing system includes a device control table 48 which stores the information proper to the communication equipments including the replacement relation in a fault mode. When a certain communication equipment receives the information showing occurrence of a fault of another communication equipment, the precedent communication equipment refers to the table 48 to retrive a replacement communication equipment. Then the connection processing is carried out by means of the retrieved communication equipment. The address information on both the faulty communication equipment and the communication equipment to be replaced are changed, and the communication processing is started again after production of an interruption.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a plurality of distributed processing computers and a plurality of communication devices connected via a LAN (local area network) line, and these communication devices connected to a wide area line network via a line switching device. The present invention relates to a communication device substitute processing method suitable for use in a data communication system.

[0002]

2. Description of the Related Art Conventionally, inter-computer communication in a distributed processing system has often taken the following form. One is a method of connecting a plurality of computers with a LAN line (local area network), and the other is a method of connecting a plurality of computers with a wide area network. There are also combinations of these.

Computer-to-computer connections such as office computers and minicomputers are made using a communication controller (communication interface) connected to the computer body. FIG. 6 shows an example of the connection.

Conventionally, the number of communication control devices that can be mounted on a computer main body is limited. Therefore, in a system in which various communication protocols exist in a multi-vendor environment, there is a limit to connection, and the processing of communication systems is limited. The CPU of the computer itself has come to be loaded.

Transfer data is increasing with the increase in scale of distributed processing systems and the increase in speed of computers. Therefore, a self-supporting communication control device is used, and the load of the communication system can be removed from the computer body. A plurality of or one computer relies on the self-supporting communication control device for the processing of the communication system, and the load can be distributed. An example of this connection is shown in FIG.

In this type of communication processing system using the self-sustaining communication control device, communication processing is intensively performed, so that the failure of the self-supporting communication control device becomes a fatal obstacle. Further, since it is difficult to integrally manage the communication control devices of the respective computers, there is a problem that communication resource maintenance efficiency is not good.

[0007]

As described above, in the communication processing system using the self-sustaining communication control device, since the communication processing is concentrated, the failure of the self-sustaining communication control device becomes a fatal failure. Become. Further, since it is difficult to integrally manage the communication control devices of the respective computers, there is a problem that communication resource maintenance efficiency is not good.

The present invention has been made in view of the above circumstances, and uses a device management table that multiplexes communication systems of a distributed processing system and manages information of communication devices associated with nodes constituting the distributed processing system. To dynamically grasp the state of communication equipment in the distributed processing system,
An object of the present invention is to provide a communication device alternative processing method capable of performing alternative processing without affecting the system even when a failure occurs.

[0009]

According to the present invention, a plurality of distributed processing computers and a plurality of communication devices are connected via a LAN (local area network) line, and these communication devices are connected to a wide area network via a line switching device. In the data communication system connected to, each of the communication devices is associated with each node that constitutes the distributed processing system,
At least contains a device management table that stores information unique to the communication device that stores the substitution relationship when a failure occurs, and when a communication device receives a failure notification of another communication device, a built-in device management table To search for a substitute communication device, perform connection processing using the communication device obtained here, change the address information of the faulty communication device and the replacement destination communication device, and issue an interrupt. It is characterized in that the communication processing is restarted. In the device management table, in addition to the substitution relationship indicating the group of communication devices that can be replaced in the event of a failure, the communication device's own address, operating status, and the transmission source when sending data to the system A sender address indicating an address, a destination address when transmitting communication data sent to the communication device, a node nickname specified in the environment generation file, and the like are stored.

[0010]

The device management table is a table for storing and managing the information of the communication device associated with each node constituting the distributed processing system. Specifically, the address of each communication device, the operating state, the relationship of substitution, the address notified to the system, the address for transmission, the nickname, etc. are stored. This table exists in the firmware of the independent communication control device, which is the core of the communication of the distributed processing system, and the communication firmware of each distributed processing computer. By using this table, it becomes possible to dynamically grasp the state of the communication device in the distributed processing system, and when a failure occurs, the alternative processing can be performed without affecting the system. That is, when a communication device receives a notification that a failure has occurred in another communication device, it refers to the built-in device management table to search for an alternative communication device in the order of registration and connects using the communication device obtained here. Do the processing
Change information such as the status of the communication device that failed, the status of the alternative communication device, and change the address information of the failed communication device and the alternative communication device considering the load on the system. Issue a lady tension and restart communication processing.

As a result, the influence of the failure can be suppressed and the throughput can be improved. Further, the substitution process can be executed without imposing any burden on the system or application.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the present invention. The distributed processing computer 11 is equipped with at least two LAN interfaces 13 for connecting to the independent communication control device 12. The self-supporting communication control device 12 is equipped with at least two LAN interfaces 14 for connecting to the distributed processing computer 11 and at least two network interfaces 15 for connecting to a wide area network. At least two distributed processing computers 11 and independent communication control devices 12 are connected to each other by a high-speed LAN 16 such as FDDI.

The LAN interfaces 14 all realize the same communication protocol. In the independent communication control device 12, at least two network interfaces 15 having the same communication protocol are mounted. Further, the network interface 15 having the same communication protocol is mounted on at least two independent communication control devices 12.

By constructing a distributed processing system in which the LAN interface 13, the network interface 15, and the self-supporting communication control device 12 are multiplexed, when a failure occurs in each communication device, an alternative process in another communication device is performed. Will be possible. Therefore, the firmware built in the self-supporting communication control device 12 and the LAN interface face firmware built in the distributed processing computer 11 include
A function for monitoring the status of each communication device is added.

The status monitoring function thus added manages the status of each communication device to select an alternative device or perform an alternative process. At this time, the device management table is used to prevent the system from being affected.

This table exists in the LAN interface firmware of the self-supporting communication controller 12 and the distributed processing computer 11. The structure of the device management table described above is shown in FIG.

The substitution process using this device management table is performed as shown in FIG. In FIG. 2, because the node nickname NCP3 has gone down,
NCP2 that is in standby state and has a high priority of substitution
Indicates that they are substituting. Here, the node nickname is the tentative name of the communication device specified in the firmware environment generation file, the address is the address of each communication device,
The operating state indicates three states of ready / hot standby / down, and the substitution relation indicates a group of substitutable communication devices. The system notification address is the sender address when sending data from this communication device to the system.
The transmission address indicates a destination address of communication data sent to this communication device. The substitution order is determined by the earliest registration in the same substitution relationship group.

FIG. 3 is a diagram showing an alternative process using the device management table shown in FIG. 2, and shows the process contents of each device in the system. FIG. 4 is a diagram showing a connection configuration example of the distributed processing system according to the exemplary embodiment of the present invention, which further embodies the configuration of FIG.

In the figure, reference numeral 41 denotes a distributed processing computer, which processes a request for application software in cooperation with other distributed processing computers or independently. Reference numeral 42 is an independent communication control device, which manages a plurality of communication interfaces.

Reference numeral 43 is a high speed LAN interface for interfacing the distributed processing computer 41 to the self-supporting communication control device 42, and realizes a communication protocol for connecting the distributed processing computer 41 to the high speed LAN line 46.

Reference numeral 44 is a high-speed LAN interface for interfacing the self-supporting communication control device 42 to the distributed processing computer 41, and realizes a communication protocol for connecting the self-supporting communication control device 42 to the high-speed LAN line 46. .

Reference numeral 45 is a network interface, which realizes a communication protocol for connecting the self-supporting communication control device 42 and the wide area network. Reference numeral 46 is a high-speed LAN line connecting the distributed processing computer 41 and the self-sustaining communication control device 42, and is realized by, for example, FDDI.

Reference numeral 47 is a line switching device, which disconnects the communication lines connecting from the plurality of independent communication control devices 42 to the wide area network to switch the lines. Reference numeral 48
Is a device management table, which is composed of a table storing the state and address information of the connection-destination communication device, and is used in the communication device replacement process, which is a feature of the present invention. Details of the device management table 48 will be described later.

FIG. 5 is a flow chart showing an operation procedure by the firmware for the alternative processing of the communication device which is a feature of the present invention. In the figure, reference numeral 51 is a step for selecting an alternative device, and the device management table 48 is searched for a communication device as an alternative device in the order of registration. Reference numeral 52 is a step of connecting the alternative device to the system, and the connection process is executed using the communication device selected in step 51. Reference numeral 53 is a step for changing the contents of the device management table 48, and changes information such as the state of the communication device in which a failure has occurred and the state of the communication device of the replacement destination. Reference numeral 54 is a step of changing the contents of the device management table 48, and changes the address information of the communication device in which the failure has occurred and the replacement destination communication device so as not to burden the system. Reference numeral 55 is a step of returning, where a lady tension is issued and communication processing is restarted.

The operation of the embodiment of the present invention shown in FIG. 4 will be described below with reference to the flow chart shown in FIG. In the embodiment of the present invention, in the distributed processing system, as shown in FIG. 4, a plurality of distributed processing computers 41 and independent communication control devices 42 are connected, and communication firmware built in each of these devices 41, 42. Will work together to improve the throughput of communication systems.

Specifically, the information transmitted from the system of the distributed processing computer 41 is the LAN interface 43 of the distributed processing computer 41, the LAN interface 44 of the self-supporting communication control device 42, the self-supporting communication control device 42, the network interface. It is transmitted to a wide area network (not shown) via 45.

Received data from the wide area network is delivered to the distributed processing computer 41 by following a route opposite to that of transmission. The communication devices corresponding to this communication path are multiplexed, and the respective states are monitored and managed by the firmware built in the self-supporting communication control device 42 and the LAN interface firmware built in the distributed processing computer 41.

The LAN firmware built in the distributed processing computer 41 and the firmware of the self-supporting communication control device 42 are registered as alternative devices when the firmware is generated. The priority of substitution is determined in the order of registration. This is illustrated by the flow chart in FIG.

Each firmware has a device management table 48 for managing communication equipment. The device management table 48 stores substitution relations, actual addresses, addresses for communicating with the system, destination addresses, operating status information, and the like. This device management table 48
The structure of is shown in FIG.

When a failure occurs in a certain communication device and is notified, the information in the device management table 48 regarding the failed device is changed by the alternative process. Specifically, the address of the transmission destination is set to the address of the alternative device, and the operating state is set to down. Further, the information in the device management table 48 regarding the alternative device is changed. Specifically, the system notification address is changed to the address of the faulty device, and the operating state is set to ready. This allows the system to communicate without being aware of any failure or replacement.

The processing mechanism according to the embodiment of the present invention described above can be widely used for constructing a system having a high throughput, which is resistant to a communication system failure in a large-scale system or a distributed processing system.

[0032]

As described above, according to the present invention, the communication system of the distributed processing system is multiplexed and the influence of the failure can be suppressed to a low level, so that the throughput can be improved. In addition, alternative processing can be executed without imposing any burden on the system or application.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a distributed processing system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of a device management table in the above embodiment.

FIG. 3 is a diagram showing an alternative process of a communication device in the above embodiment.

FIG. 4 is a block diagram further embodying the configuration of the distributed processing system shown in FIG.

FIG. 5 is a flowchart showing a procedure of an alternative processing operation of the communication device in the above embodiment.

FIG. 6 is a diagram showing an example of a communication line connection of a distributed processing system.

FIG. 7 is a diagram showing an example of connection of a distributed processing system communication line using an independent communication control device.

[Explanation of symbols]

11, 41 ... Distributed processing computer, 12, 42 ... Independent communication control device, 13, 14, 43, 44 ... LAN interface, 15, 45 ... Network interface, 16, 4
6 ... High-speed LAN line, 17, 47 ... Line switching device,
48 ... Device management table.

Continuation of front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location H04L 12/437

Claims (2)

[Claims] 1. A plurality of distributed processing computers and a plurality of communication devices are LANs (local area networks).
In a data communication system connected via a line and connecting these communication devices to a wide area line network via a line switching device, at least each communication device is associated with each node constituting a distributed processing system, It has a built-in device management table that stores information unique to the communication device that stores the substitution relationship when a failure occurs, and when each communication device receives a notification of the failure occurrence of another communication device, the built-in device management table The communication device to be referred to is searched for, the connection process is performed using the communication device obtained by this search, the address information of the failed communication device and the communication device of the replacement destination is changed, and an interrupt is issued to communicate. Communication equipment alternative processing method characterized by restarting processing. 2. In the device management table, in addition to a substitution relationship indicating a group of communication devices that can be replaced when a failure occurs, an address unique to the communication device, an operation state, a source address of data, 2. The communication device alternative processing method according to claim 1, wherein the destination address and the node nickname specified by the environment generation file are stored.