JPH07107076A - No-hit data switching device - Google Patents

Abstract

PURPOSE:To perform redundancy switching of data in a system for which high reliability is requested without losing the data. CONSTITUTION:The data loss detection part 40 of an address control part 4 monitors the counting up of addresses in the address counters 12 and 22 of data storage parts 1 and 2 and judges data loss conditions. Address backdating parts 41 and 42 backdate current addresses from the address counters 12 and 22 corresponding to fault information from OR gates 43 and 44, generate the addresses before a fault is generated and output them to data read control parts 13 and 23. A changeover control part 45 generates a changeover instruction corresponding to backdating information from the address backdating parts 41 and 42 and outputs the changeover instruction to a data switching part 3.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a hitless data switching device, and more particularly to a serial data switching device.

[0002]

2. Description of the Related Art Conventionally, in a switching device of this type, as shown in FIG. 2, serial or parallel data and data are switched by a data switching unit 6. here,
The data switching unit 6 uses a mechanical relay or an electronic relay.

An instruction to switch data in the data switching section 6 is controlled by an external switching control signal. For example, when an external switching control signal is input while data is selected by the data switching unit 6, a mechanical relay or an electronic relay operates in the data switching unit 6 to switch to data.

[0004]

In the above-mentioned conventional switching device, data is switched by a mechanical relay or an electronic relay, so that continuous data always suffers data loss during the switching sequence. There is.

Further, since the data switching is performed by the external switching control signal after detecting the data loss,
There is a problem that the time of data loss becomes long and the meaning of making data redundant in a system that requires high reliability is halved.

Therefore, an object of the present invention is to solve the above problems and provide a non-interruptive data switching device capable of performing redundant switching of data in a system requiring high reliability without data loss. .

[0007]

A hitless data switching device according to the present invention comprises first and second storage means for converting serial data from the outside into parallel data and storing the parallel data, and the first and second storage means. Switching means for selecting and outputting parallel data from one of the storage means, and the first and second storage means when a failure of the serial data to one of the first and second storage means is detected. Based on the address designated by the address before the failure occurs in the other of the second storages, and the other of the first and second storages based on the address designated by the address designator. Switching instruction means for instructing the switching means to select the parallel data output from.

[0008]

An embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, the serial-parallel conversion units 10 and 20 of the data storage units 1 and 2 perform serial-parallel conversion of the input data and execute the memory 11 and 2.
Output to 1. The memories 11 and 21 temporarily store the parallel data sent from the serial-parallel converters 10 and 20.

The address counters 12 and 22 are serial-
Each time the serial-parallel conversion is performed by the parallel conversion units 10 and 20, the count-up is performed, and the count value is used as an address, the data read control units 13 and 23, the data loss detection unit 40 and the address backdate unit 41 of the address control unit 4, To 42.

The data read control units 13 and 23 have address counters 12 and 22 and an address back date unit 41,
Based on the address from 42, reading of parallel data from the memories 11 and 21 is controlled.

The data switching unit 3 is responsive to a switching instruction from the switching control unit 45 of the address control unit 4 to store data in the data storage units 1 and 2.
The parallel data is switched from the memories 11 and 21 and the selected parallel data is output to the data output unit 5.

A data loss detecting section 40 of the address control section 4
Detects the data loss in the data input to the data storage units 1 and 2 based on the current addresses from the address counters 12 and 22 of the data storage units 1 and 2. That is, the data loss detection unit 40 uses the address counter 1
Monitor the count-up of addresses at 2, 22,
Determine the data loss situation.

The address backdate units 41 and 42 backdate the current addresses from the address counters 12 and 22 according to the fault information from the OR gates 43 and 44, and the backdated addresses are read from the data read control unit 1.
It outputs to 3,23.

That is, the address backdate unit 41,
Reference numeral 42 generates an address before failure occurrence in the data storage units 1 and 2 by backdate of the current address from the address counters 12 and 22, and outputs the address before failure occurrence to the data read control units 13 and 23.

The OR gates 43 and 44 take the logical sum of the data loss detection signal from the data loss detection unit 40 and the fault status from the outside, and output the operation result as fault information to the address backdate units 41 and 42.

The switching control unit 45 generates a switching instruction according to the backdate information from the address backdate units 41 and 42, and outputs the switching instruction to the data switching unit 3. That is, the switching control unit 45 uses the address backdate unit 41,
When the information indicating that the backdating is performed is input from 42, the data switching unit 3 is switched to switch to the parallel data from the data storage units 1 and 2 corresponding to the address backdating units 41 and 42 that performed the backdating. Output the switching instruction.

The data output unit 5 converts the parallel data from the data storage units 1 and 2 selected by the data switching unit 3 into serial data, and outputs the serial data (SELECT).
EDDATA) is output to the outside.

The operation of the embodiment of the present invention will be described in detail with reference to FIG. Hereinafter, a case where data loss occurs in the data input to the data storage unit 1 while the data switching unit 3 is selecting the parallel data from the data storage unit 1 will be described.

The data loss detection section 40 detects that the address counter 12 has stopped counting up at address 100, and if it judges that the data is lost, the data loss detection signal is sent to the address backdate section 42 via the OR gate 44. Output.

At this time, if the address counter 22 continues counting up without being stopped, the address backdate unit 42 determines that
Backdate the current address from 2 and output the address before the data loss to the data read control unit 23.

In this case, if the address counter 22 outputs the address 100, as in the case of the address counter 12, the address backdate unit 42 backdates the data three addresses before the address 100, and the address backdate unit 42 outputs 100.
The third address is subtracted from the address and the 97th address is output to the data read control unit 23.

When the switching control unit 45 learns from the information from the address backdate unit 42 that the data from the data storage unit 2 has been backdated to the previous data, the data switching unit 3 stores the data stored in the data storage unit 1. It is instructed to switch from the data to the data in the data storage unit 2.

The data read control unit 23 is the address from the address backdate unit 42 before the data loss 97.
When the address is input, the data is controlled to be read from the address 97 of the memory 21 and output to the data switching unit 3.

In response to an instruction from the switching control unit 45, the data switching unit 3 operates so as to select the data from the data storage unit 2 when outputting the data at address 97 to the data output unit 5.

As a result, even if the data input to the address 100 of the data storage unit 1 is lost, the data input to the data storage unit 2 from the address 97 before the data loss is transferred to the data switching unit 3. Will be selected.

In the processing operation described above, the data input to the data storage units 1 and 2 in the data loss detection unit 40,
This is a processing operation when a data loss is detected in. On the other hand, even when a failure is detected by the serial data from the data output section 5 and the failure status is input to the address control section 4 from the outside, the data read destination is the data switching section 3. Can be switched. The control in that case is the same as the above-mentioned processing operation, and therefore its explanation is omitted.

If no data loss or failure occurs, the data in the data storage unit 1 or the data in the data storage unit 2 selected by the data switching unit 3 is continuously output to the outside.

As described above, the two systems of serial data are converted by the serial-parallel converters 10 and 20 and then temporarily stored in the memories 11 and 21. Before the occurrence of the fault generated by the address backdate units 41 and 42, when the fault is detected by the data loss detecting unit 40 or the fault due to the data to the data storage units 1 and 2 is notified by the fault status from outside Data loss is automatically detected by switching the data read from the normal data storage units 1 and 2 to the data read out from the normal data storage units 1 and 2 by the data switching unit 3 and reading the data in the lost period from the other system. Can be sent out. Therefore,
Redundant switching of continuous data in a system that requires high reliability can be performed without data loss.

[0030]

As explained above, according to the present invention, serial data from the outside is converted into parallel data and temporarily stored in the first and second storage means, and the first and second storage means are stored. When a serial data failure to one of them is detected, the other of the first and second storage means is instructed of the address before the failure occurs, and the first and second addresses are based on this instructed address. By switching to the parallel data output from the other of the second storage means, it is possible to perform redundant switching of data in a system requiring high reliability without data loss.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a conventional example.

[Explanation of reference numerals] 1, data storage unit 3 data switching unit 4 address control unit 5 data output unit 10, 20 serial-parallel conversion unit 11, 21 memory 12, 22 address counter 13, 23 data read control unit 40 data loss Detection unit 41, 42 Address back date unit 45 Switching control unit

Claims (3)

[Claims] 1. A first and second storage means for converting serial data from the outside into parallel data and storing the parallel data, and one of the first and second storage means for selecting and outputting the parallel data. The switching means and the failure of the serial data to one of the first and second storage means before the failure of the other of the first and second storage means occurs. Address instructing means for instructing the address, and instructing the switching means for selecting the parallel data output from the other of the first and second storage means based on the address instructed by the address instructing means. A non-instantaneous-interruption-data switching device having a switching instruction means for performing the switching. 2. The address instructing means sets an address before the occurrence of a failure in the first and second storage means in accordance with at least one of detection of data loss in the serial data and failure information from the outside. 2. The hitless data switching device according to claim 1, wherein the device is configured to instruct the other. 3. The method according to claim 1, further comprising a determining unit that determines the input state of the serial data to the first and second storage units and detects a data loss in the serial data. 2. No interruption data switching device described in 2.