JPH08186618A - Switching device for information processor - Google Patents

Abstract

PURPOSE: To take over a processing when a fault occurs to the other device through the use of all information processors by detaching the information processor where the fault occurs from a communication network and permitting the other information processor to execute the processing. CONSTITUTION: It is inquired to server computers 10-13 becoming monitored objects by the supply of the power of a switching device 20 whether they are normally operated in order. When it is ascertained that the fault occurs in the server computer 10 and it does not normally operate, the address '110' of the server computer 10 is stored, the server computer 10 is removed from the monitored object and a switch part 22 is opened (turned off) based on the stored address '110' for detaching the server computer 10 where the fault occurs from the communication network 1. An instruction for making the other normal server computers 11-13 to take over a processing request for the server computer 10 is given to an address control part 26.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a switching device connected to an information processing device, and more particularly, when a failure occurs in the information processing device connected via a communication network, the processing is transferred to another information processing device. It is related to technology to take over and continue.

[0002]

2. Description of the Related Art Conventionally, in a system that emphasizes reliability,
Local Area Network (Local Area)
As seen in a client / server system using a network (LAN), another information processing device is prepared as a standby system in the event of a failure in the information processing device serving as a server, and the active / standby system is used. It has a composition.

In the above system, there are the following two methods as a method of handing over the processing from the active information processing apparatus to the standby information processing apparatus.

(1) The standby system information processing device takes over the address assigned to the active system information processing device and continues processing.

(2) The information processing apparatus in which a failure has occurred, using the broadcast method, notifies the information processing apparatus that issues the processing request of the address change of the processing request destination. After that, the information processing apparatus on the processing request side issues a processing request to the changed address to continue the processing.

However, in these methods, when a failure occurs in the information processing apparatus of the active system, in the method (1), the information processing apparatus on the processing execution side is processed, and in the method (2), the information processing on the processing request side is processed. Each device had to bear the address change process.

As a measure for improving both of the above methods, the processing request side,
JP-A-2-15913 discloses a method in which neither the information processing apparatus on the processing execution side is burdened with the address change processing.
A typical method disclosed in Japanese Patent Publication No. 9 is to switch the information processing apparatuses of the active system and the standby system by a changeover switch. In this switching method using the changeover switch, when a failure occurs in the information processing device of the active system, first, by switching the changeover switch, the information processing device of the active system is disconnected from the communication network and the information processing device of the standby system is connected. To do.

Next, the address conversion section of the change-over switch converts the address information contained in the data addressed to the failed information processing apparatus into the address information addressed to the standby information processing apparatus.

On the other hand, the address information addressed to the address conversion unit, which is included in the data processed by the standby information processing device, is converted into the address information of the active information processing device by the address conversion unit.

[0010]

The present inventor has found the following problems as a result of examining the above-mentioned prior art.

To the changeover switch, two units, an active information processing device and a standby information processing device, are simultaneously connected,
Depending on whether or not a failure has occurred in the active information processing device, the above 2
Only one of the information processing devices connected to each other is connected to the communication line.

Therefore, there is a problem that a part of the information processing device connected by the communication line cannot be used as a standby system.

An object of the present invention is to utilize all the information processing devices connected by a communication line, and when a failure occurs in the information processing device, perform the processing performed by the failed information processing device. An object is to provide a switching device of an information processing device that can be taken over by the information processing device.

The above object and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

[0015]

Of the inventions disclosed in the present application, a representative one will be briefly described below.
It is as follows.

A switching device for allowing another information processing device to take over and continue the processing executed by the information processing device having a failure among the plurality of information processing devices connected to the communication network via the changeover switch. There
Monitoring means for monitoring whether each information processing device is operating normally, disconnection means for disconnecting the information processing device in which a failure has occurred from the communication network based on the monitoring result of this monitoring means, and information processing in which the failure has occurred The address management table in which the address information of the information processing device that is the takeover destination of the processing performed by the device is registered in advance, and the address information of the data addressed to the information processing device that is disconnected from the communication network due to a failure is converted according to the address management table. And an address converting means for performing the same.

[0017]

According to the above-mentioned means, the backup information indicating which information processing device can take over the processing when a failure occurs in the information processing device is registered in advance in the address management table.

The monitoring means constantly monitors whether or not the information processing device is operating normally by inquiring sequentially to the information processing device. As a result, when it is detected that a failure has occurred in the information processing device, the address translation means and the disconnection means are instructed to disconnect.

Then, the disconnecting means to which the faulty information processing device is connected is turned off, and the faulty information processing device is disconnected from the communication network.

In the address information conversion means, the address of the transmission destination is set to the address set in the address management table, and the address of the transmission source is set to the address conversion means. It is converted into each address and sent to the communication network.

As a result, the data after the address conversion is processed by the information processing device having the address set in the address management table.

As described above, the information processing apparatus is connected to the communication network through the disconnecting means, and the monitoring means operates the disconnecting means to disconnect the failed information processing apparatus from the communication network. By causing another information processing apparatus to execute the processing performed by the generated information processing apparatus by replacing the address information by the address conversion means, the processing can be used even when the information processing apparatus fails. ,
It is possible to prevent an increase in the burden on the information processing device that takes over the processing that the information processing device that has failed is performing.

[0023]

EXAMPLES The structure of the present invention will be described below with reference to examples.

FIG. 1 is a diagram showing a schematic configuration of an embodiment of a distributed processing system to which the switching device of the present invention is applied.

In FIG. 1, 1 is a communication network and 1
Reference numerals 0, 11, 12, and 13 denote server computers (information processing devices on the processing execution side), and the addresses (AD) on the communication network 1 are "110" and "11", respectively.
1 ”,“ 112 ”, and“ 113 ”.

Reference numeral 20 denotes a switching device for disconnecting from the communication network 1 a health check unit 21 for monitoring whether the server computers 10 to 13 are operating normally and a server computer in which a failure has occurred according to an instruction from the health check unit 21. Changeover switch units 22, 23, 2
4, 25, and client computers (information processing devices on the processing request side) 30, 31 and server computers 10 to 13 in order to cause another server computer to execute the processing performed by the failed server computer.
And an address control unit 26 for switching the address of communication data between them.

The client computers 30 and 3
The addresses on the communication network 1 of No. 1 are “13
0 ”and“ 131 ”, and the address of the address control unit 26 is“ 126 ”.

This address control unit 26 is further provided in the case where any of the server computers 10 to 13 fails.
It is composed of an address management table 27 for pre-registering address information of a server computer that takes over the processing, and an address conversion unit 28 for converting address information of data addressed to the failed server computer according to the address management table 27. It

Further, the server computers 10 to 13 are in a standby state for the processing executed by the server computer which will take over the processing based on the address management table 27, in order to speedily carry over the processing when a failure occurs. .

That is, in the address management table 27,
When the takeover destination of the server computer 10 is designated as the server computer 11, different processing programs should originally be executed, but in case the server computer 10 fails, the server computer 11 has a server. By keeping the processing program executed by the computer 10 in the standby state,
By switching over when a failure occurs, the succeeded processing can be performed quickly.

FIG. 2 is a schematic configuration diagram of the changeover switch units 22 to 25.

In FIG. 2, reference numeral 41 is an input / output terminal connected to the server computers 10 to 13, and 42 is an input / output terminal connected to the communication network 1.

Reference numeral 43 is a switch for the health check unit 21.
ON / O by control signal input 44 connected to
FF is performed to control the connection between the server computers 10 to 13 and the communication network 1.

Reference numeral 45 is an input / output terminal connected to the address control unit, and is connected to the input / output 42 connected to the communication network 1 by a wired OR.

FIG. 3 is a table configuration diagram of the address management table 27, and 51 and 52 are addresses "110" and "111" of the server computers 10 and 11 at the time of normal operation.
53 and 54 have addresses “110” and “11”.
When a failure occurs in the server computer 10, 11 of 1 ”, the server computer 11, 1 that takes over the processing
2 addresses “111” and “112”.

That is, in FIG. 3, when the server computer 10 of the address "110" fails, the server computer 11 of the address "111" takes over the processing, and the server computer 11 of the address "111" fails. The case indicates that the server computer 12 having the address “112” takes over the processing.

FIG. 4 is a schematic diagram of the data format of the LAN of the embodiment of FIG. 1, where 61 is a receiver address (destination address) area, 62 is a data area, and 63 is a transmitter address (source address) area. .

Next, FIG. 5 shows a schematic operation algorithm of the health check unit 21 of the switching device 20 in the distributed processing system of this embodiment. For example, the operation of the switching device 20 when a failure occurs in the server computer 10 is shown in FIG. It will be described based on 5.

When the switching device 20 is turned on, the algorithm starts. First, in step 501, the server computers 10 to 13 to be monitored are inquired in order whether or not they are operating normally, and then the next step 5
Go to 02.

In step 502, it is determined whether the server computers 10 to 13 are operating normally based on the result of step 501.

As a result, all server computers 10
If ~ 13 is operating normally, step 50 again.
Returning to 1, but the server computer 10 has failed,
If it is determined that the operation is not normal, the process proceeds to the next step 503.

In step 503, the address “110” of the server computer 10 in which the failure has occurred is stored, and the server computer 1 is monitored from the monitoring target in step 501.
Except for 0, the process proceeds to the next step 504.

In step 504, the address "110" stored in step 503 is set in order to disconnect the faulty server computer 10 from the communication network 1.
The switch unit 22 is opened (OFF) based on the above, and the process proceeds to the next step 505.

In step 505, the address controller 26
To the other normal server computers 11 to 13, and then returns to step 501.

Next, FIG. 6 shows a schematic operation algorithm of the address control unit, and the operation of the address control unit instructed by the health check unit 21 to take over will be described based on the algorithm of FIG.

The instruction from the health check unit 21 to take over the process upon detection of a failure is the start of this algorithm. First, in step 601, the data existing on the communication network 1 is referred to, and the flow advances to step 602.

In step 602, the data referred to in step 601 is the server computer 1 in which the failure has occurred.
Data addressed to 0, that is, the transmitter address is “11
It is determined whether the data is "0", and if the data is of a different destination, the process proceeds to step 605.

On the other hand, if the data is the corresponding data, the data receiving device address 61 addressed to the server computer 10 "110" is the data receiving address 61 addressed to the server computer 11 based on the address management table 27 in step 603. Convert to "111", step 6
Go to 04.

In step 604, the data transmission device address 63 "130" of the client computer 30 which has issued the data processing request is converted into the address "126" of the address control unit 26, and the process returns to step 601.

On the other hand, steps 602 to 605
If the process proceeds to step 1, first, it is determined whether the address is the address addressed to the address control unit 26, and if the address is different, step 601.
Return to

However, if it is addressed to the address control unit 26, the process proceeds to step 606.

In step 606, the data receiving address 61 addressed to the address control unit 26, "126", is converted into the address "130" of the client computer 30 which issued the data processing request, and the flow advances to step 607.

In step 607, the address "111" of the server computer 11 that has taken over the process in place of the failed server computer 10 is converted into the address "110" of the server computer 10, and the process returns to step 601. .

As described above, according to this embodiment, the changeover switches 22 to 25 are used to switch the server computers 10 to 13 respectively.
The health check unit 21 operates the changeover switch 22 to disconnect the faulty server computer 10 from the communication network 1, and the server computer 10 performs the operation. By causing another server computer 11 to execute the above processing by replacing the address information by the address control unit 26, the processing of the failed server computer 10 can be taken over by the server computer 11 without increasing the load at the time of switching. The effect is that you can do it.

As described above, the invention made by the present inventor is
Although the present invention has been specifically described based on the above-mentioned embodiments, the present invention is not limited to the above-mentioned embodiments, and it goes without saying that various modifications can be made without departing from the scope of the invention.

For example, server computer 10 to 1
4 of 3 and 2 of client computers 30 and 31
Although the description has been given based on the embodiment of FIG. 1 using a single computer, the information processing device switching device of the present invention can be applied to a system having two or more server computers and one or more client computers. Needless to say.

[0057]

The effects obtained by the typical ones of the inventions disclosed in the present application will be briefly described as follows.

(1) A plurality of information processing devices connected to a communication line can be used as active system information processing devices without distinction between active and standby systems.

(2) When a failure occurs in the information processing apparatus, the processing performed by the failed information processing apparatus is replaced by another information processing without burdening either the processing request source or the processing request destination. Can be taken over by the device.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a distributed processing system that is an embodiment of the present invention.

FIG. 2 is a schematic configuration diagram of a changeover switch unit.

FIG. 3 is a table configuration diagram of an address management table.

FIG. 4 is a schematic diagram of a LAN data format of the distributed processing system according to the present embodiment.

FIG. 5 is a diagram showing a schematic operation algorithm of a health check unit.

FIG. 6 is a diagram showing a schematic operation algorithm of an address control unit.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Communication network, 10, 11, 12, 13 ... Server computer, 20 ... Switching device, 21 ... Health check part, 22, 23, 24, 25 ... Changeover switch part, 26 ... Address control part, 27 ... Address management table , 28 ... Address conversion unit, 30, 31 ... Client computer.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical indication H04L 12/40

Claims (1)

[Claims] 1. A switch for continuing processing of another information processing device, which was being executed by a faulty information processing device among a plurality of information processing devices connected to a communication network via a changeover switch, to another information processing device. A monitoring unit that monitors whether or not each information processing device is operating normally; a disconnection unit that disconnects the information processing device in which a failure has occurred from the communication network based on the monitoring result of this monitoring unit; The address management table in which the address information of the information processing device that is the takeover destination of the processing performed by the information processing device in which the error occurred is registered in advance, and the address information of the data addressed to the information processing device that has been separated from the communication network due to the failure An information processing device characterized by comprising an address conversion means for converting according to an address management table. Switching device.