JPH0159605B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a duplex file updating method that improves file reliability.

In processing systems, one way to increase file reliability is to use dual volumes.
There is. For example, important data that is unnecessary for restarting processing after an accident, such as check information obtained by the checkpoint restart function, is protected by duplicating files. This method has the advantage that even if one of the files is damaged, the data in the other file can be used. However, although this conventional method is effective against a failure of a file device or a channel device, it is not necessarily effective against a power outage due to a power outage accident. That is, if power is cut off while the check information is being written to two file devices at the same time, both data files of the check information will be damaged. As a preventive measure, a method has been adopted to prevent file damage by preparing a DC power source such as a generator or batteries for use during power outages. As is well known, the power supply equipment for power outages is expensive and requires a large installation space. Therefore, in medium to small-sized processing systems, work management processing systems installed at work sites, etc., the above-mentioned equipment makes the entire system expensive and there are drawbacks such as being subject to restrictions on the installation location.

The present invention has been made to solve the above-mentioned drawbacks, and aims to provide a file update method for a duplicated file that enables duplication of important data files without requiring power supply equipment for power outages.

The present invention provides a duplex file control method that includes a processing device, a storage device, and two file devices, and stores necessary data in the storage device in the two file devices. identification code adding means for adding the same identification code to the beginning and end of the required data; and after storing the required data to which the identification code has been added in one of the two file devices, the data is stored in the other file device. and an identification code inspection means for comparing and inspecting an identification code added to the beginning and an identification code added to the end of the data stored in the file device. This is a file update method for duplicated files.

Further, the present invention is provided with means for measuring time, and means for adding the same time measurement data by the time measurement means to the beginning and end of the required data stored in the file device.

Hereinafter, the present invention will be explained with reference to the drawings. The drawing is a block diagram illustrating an embodiment of the present invention.
1 is an input device, 2 is a main storage device, 3 is a processing device,
4 is a control unit, 5 is a processing unit, 6 is a check unit, 7 is a time counter, 8 and 9 are file devices, 10 is an output device, A ₁ and A ₂ are files, D is data, T ₁ ,
T ₂ is time measurement data, b is the leading part, and C is the trailing part. In the drawing, it is assumed that the output of the time counter 7 is time data of year, month, day, hour, minute, and second, and the check section 6 is assumed to have a check point restart function. In the drawing, input data from an input device 1 is processed by a processing device 3, and the result is outputted to an output device 10. During the execution of the processing, the control section 4 activates the check section 6 to check the execution process. The required data is stored in the main storage device 2. That is, the check section 6 collects data from the instruction register, address register, etc. at predetermined time intervals (t ₁ , t _{2 .} . . ) and stores the collected data in the main memory 2 . Therefore, the data D is check information, which is stored in the file devices 8 and 9. For example, when a processing system failure occurs at time _t2 , the check information at time _t1 in data D can be used. Therefore, the processing after time t ₁ can be resumed continuously. Conventionally, data D was stored in file devices 8 and 9.
When storing data, they were done simultaneously and in parallel. Therefore, if a power outage occurs during storage, both files will be damaged. To prevent this, power supply equipment for power outages is required as described above, but the present invention eliminates the need for power supply equipment by storing data D in two file devices in series. For example, after storing data D in the file device 8, the same data D is stored in the file device 9.
Store in. Furthermore, the time data _T1 (year _{, month} ,
Day, hour, minute, second) are added and stored. By performing serial storage in this way, for example, at time t ₂ while data D is being stored in the file device 8,
Even if a power outage accident occurs, data D (time t ₁ ) of the file device 9 can be used, so processing can be resumed after time t ₁ . In addition, if an accident occurs during storage of data D, the file in file device 8
By comparing and determining the clock data T ₂ (time t ₂ ) of the leading part b of A ₁ and the clock data T ₁ (time t ₁ ) of the tail part c of A 1, it is easy to determine that file A ₁ is incorrect. can be detected. On the other hand, file device 9
Since the time measurement data of the first part b and the last part c of the file _A2 are both _T1 , the data D in this file _A2 can be used as check information for restarting the process. Note that the present invention is applicable not only to the above-mentioned file where 1 file = 1 logical record, but also to files where 1 file = n logical records.

As described above, the present invention has the advantage that even in a processing system that is not equipped with a power supply for power outages, it is possible to prevent the effects of power outage accidents on redundant files.

[Brief explanation of drawings]

The drawing is a block diagram illustrating an embodiment of the present invention, and the symbols used in the drawing are as follows. 1 is an input device, 2 is a main storage device, 3 is a processing device, 4 is a control section, 5 is a processing section, 6 is a check section,
7 is a clock counter, 8 and 9 are file devices, 10
is the output device, A ₁ A ₂ is the file, D is the data,
T ₁ T ₂ indicates time measurement data, b indicates the leading portion, and c indicates the trailing portion.

Claims (1)

[Scope of Claims] 1. In a duplex file control system comprising a processing device, a storage device, and two file devices, and in which necessary data in the storage device is stored in the two file devices, identification code adding means for adding the same identification code to the beginning and end of the stored required data; and after storing the required data to which the identification code has been added in one of the two file devices, the other is serial storage control means for storing the data in the file device; and identification code inspection means for comparing and inspecting the identification code added to the beginning and the identification code added to the tail of the data stored in the file device. A file update method for a duplicated file, which is characterized by the following. 2. The processing device is equipped with a timer for measuring time, and the identification code adding device is configured to add clock data by the timer as an identification code to the beginning and end of the required data stored in the file device. A file update method for a duplicated file according to claim 1, characterized by this.