JPS61156438A - Computer control device - Google Patents

Abstract

PURPOSE:To execute the back-up action even when either of back-up computers breaks down by selecting either of them with a selecting circuit at the time of the back-up action with regard to the self diagnosis result of double back-up computers. CONSTITUTION:When a computer 1 executes the detection by a self diagnosis processing, the computer outputs a self diagnosis 51 to a monitoring circuit 31 and outputs a change-over signal 71. A monitoring circuit 102, when the change-over signal 71 is inputted, stops the output of a change-over permitting signal 106. The change-over permitting signal 106, when both of computers 100 and 101 break down, namely, both self diagnosis signals are outputted, stops the output. Thus, when both computers 100 and 101 break down or while either of them is backed up, a monitoring circuit 102 executes the change-over disable for other computers 22-2n, further, the monitoring circuit 102 circuit 102 outputs a selecting signal 111 to select a broken-down computer 21 in accordance with the change-over signal to the computers 100 and 101, and executes the back-up processing to the computer 21.

Description

[Detailed description of the invention] [Industrial application field] Regarding.

[Prior Art] Conventionally, in a computer control system in which a plurality of computers are distributed, when each computer fails, a backup computer I is usually installed to execute processing in place of the failed computer.
I is provided. This backup computer will perform the processing that should be executed by that computer, depending on the result of the self-diagnosis processing of the 1 computer that has failed.

[Problems to be Solved by the Invention] Incidentally, the above-mentioned backup processing has conventionally been executed by one backup computer. Therefore, if the backup computer itself breaks down, backup processing becomes impossible, resulting in a problem of lowering the reliability of the entire system.

An object of the present invention is to provide a computer control device that can improve the reliability of a backup system by duplicating backup computers, thereby increasing the reliability of the entire computer system.

[Means and effects for solving the problems] In the present invention,
Duplicate backup computers are provided, and one of them is selected by a selection circuit during backup operation based on the self-diagnosis results of the duplex backup computers. While the backup operation is being performed by the redundant backup computer, the blocking circuit blocks the backup operation for computers other than the failed computer.

With a backup system having such a configuration, even if one of the backup computers fails, the backup operation can be executed, and reliability can be improved.

[Example] An example of the present invention will be described below with reference to the drawings. 1st
The figure is a block diagram showing the configuration of a calculation control system according to an embodiment. In FIG. 1, n calculations tI2
1 to 2n are controlled objects (
control signals 61 to 6n (not shown) are output. When the computers 21 to 2n detect that a failure has occurred through self-diagnosis processing, they output self-diagnosis signals 51 to 5n. The monitoring circuits 31 to 3n output switching signals 71 to 7n based on the respective self-diagnosis signals 51 to 5n and the switching permission signal 106 from the monitoring circuit 102. The switches 41 to 4n select one of the control signals 61 to 6n or the output signal 110 from the switch 103 according to each of the switching signals 71 to 7n, and output signals 81 to 80.

On the other hand, calculations 1ioo and iol are backup computers, to which the input signals 11 to 1n are input, and the selection signal 111 from the monitoring circuit 102 is inputted.
Perform backup processing accordingly. Also, calculation t11
00 and 101, when a failure is detected through self-diagnosis processing, self-diagnosis signals 104 and 105 are sent to the monitoring circuit 1.
Output to 02. The monitoring circuit 102 outputs the self-diagnosis signal 10
4, 105 and the switching signals 11 to 1n, a switching permission signal 106 and a switching signal 107 are output.

The switching device 103 switches the calculation screen 100 according to the switching signal 107.
One of the control signals 108 and 109 from 101 is selected and outputted as an output signal 110.

In the computer control system having such a configuration, the operation of this embodiment will be explained. First, if each computer 21 to 2n is not malfunctioning and is operating normally, each computer 21 to 2n
The control signals 61 to 6n output from the switches 41 to 6n
4n and sent to the control pair Φ as output signals 81 to 8n.

Next, let us assume that, for example, the calculation lI21 has failed. Calculation 1!
1121 outputs a self-diagnosis signal 51 to the monitoring circuit 31 when a failure is detected through self-diagnosis processing. Monitoring circuit 31
As shown in FIG. 2, it consists of an AND circuit 20, and outputs a switching signal 71 when a self-diagnosis signal 51 and a switching permission signal 106 are input.

Here, the monitoring circuit 102 includes an OR circuit 90. as shown in FIG. It consists of a logic circuit including a NOT circuit 91 and an AND circuit 92. That is, the monitoring circuit 102 receives the switching signal 7
When 1 is input, the output of the switching permission signal 106 is stopped. This switching permission signal 106 calculates tg1100,
101 is out of order, that is, the self-diagnosis signal 10
When both 4 and 105 are output, the output is stopped. As a result, if both the computers 100 and 101 are out of order or one of them is being backed up, the monitoring circuit 102 disables switching to the other computers 22 to 2n. Further, the monitoring circuit 102 outputs a selection signal 111 to the computers 100 and 101 to select a computer 21 that has failed in response to the switching signal 71. This allows calculation 1
1100 and 101 perform backup processing for the computer 21 that performs processing according to the input signal 11. Through this backup process, control signals 108 and 109 are output from the computers 100 and 101 to the switch 103.

The monitoring circuit 102 generates a switching signal 107 in response to self-diagnosis signals 104 and 105 from the computers ioo and iol.
is output to the switch 103. Here, the switching signal 107 is a signal that selects calculation 11i1101 when the computer 100 fails, and selects calculation 1ioo in other cases.

The switching device 103 receives a control signal 108 output from the computers 100 and 101 in response to the switching signal 101 as described above.
Or select one of the control signals 109 and output the output signal 1.
10 is output to the switch 41. The switch 41 selects the output signal 110 of the switch 103 in response to the switch signal 71 and outputs the selected signal to the controlled object.

In this way, my phone broke down! Instead of the control signal 61 of 1121, one of the control signals of calculation va100 and 101 is selected by the switch 41 and sent to the controlled object. In addition, even if a total of n machines other than calculation 11121 fail, Ht H+su100,1
01 is executed.

By the way, in the present invention, backup processing is executed by the backup computers 100 and 101 which are duplicated as described above. Here, if the failure rate of a computer is λ and the repair time is the king, the unreliability U per computer is expressed by the following formula (1).

U=λT/(1+λT)...(1) Furthermore, the unreliability ub of the entire duplicated backup computer is as follows: ub=uz...(2). If U<1, then LJb<U, so the unreliability ub of the entire duplicated backup computer will be a small value. Normally, since U is 1, the unreliability ub is so small that it can be ignored compared to U.
The reliability of the duplicated backup computer is extremely high. For example, λ is 10-'/h, T is 10h
Then, U is io゛3. ub becomes 101, and the unreliability can be reduced to 1/1000.

[Effects of the Invention] As described in detail above, according to the present invention, in a computer control system, the reliability of backup processing for a failed computer can be greatly improved by using a redundant backup computer system. Therefore, the reliability of the entire computer control system can be improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of a computer control system according to an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of monitoring circuits 31 to 3n of the embodiment, and FIG. 2 is a block diagram showing the configuration of a monitoring circuit 102. FIG. 21-2n...computer, 31-3n...monitoring circuit,
41-4n...Switching device, 100, 101...Backup computer, 102...Monitoring circuit, 103...
・Switcher.

Claims (1)

[Claims] In a distributed computer system of n units, there is a redundant backup computer that executes the arithmetic processing of that computer when one of the computers fails, and a self-diagnosis result of this redundant backup computer. a selection circuit that selects one of the duplex backup computers based on the duplex backup computer; and a blocking circuit that prevents backup operations for computers other than the failed computer while the duplex backup computer is performing a backup operation. It is equipped with a computer control device.