JPS6121539A - Generation management system of data file - Google Patents

Abstract

PURPOSE:To attain processing of restoration at re-application of power without any error by providing a generation recognition code area to plural parts of a record block and renewing a code of the generation recognition code area every time data are written. CONSTITUTION:When data from a main memory is outputted to a bus (e), a bus interface control section 5 of a file control channel designated to the control section fetches the data and stored once in a memory RAM9 by the control of a microprocesor MUP6. Then a prescribed data file is started by a file interface control section 10 according to the control program stored in a memory ROM7, the data in the RAM9 is transferred to a data file by the command of the MPU6 and it is written and renewed. If power supply interruption takes place during on the way of writing, the write renewal of the transfer data is intemitted. When the power is turned on again, the code data of the generation recognition code in the read data is compared by the MPU6 and if any difference is detected, the data transfer is requested newly and the write of the transfer data is tried again.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a generation management system for record blocks composed of a plurality of sectors.

With the spread of data processing, the amount of data input and output by terminal devices constituting information processing systems has rapidly increased, and as a result, demand for small, simple, low-cost, and easy-to-handle terminal devices has increased.

Furthermore, with the spread of minicomputers and microcomputers, the functions of information processing systems have been distributed, and terminal devices have come to be used not only as data input/output devices but also as data creation devices.

One method of recording and managing data handled by such a terminal device in a data file within the terminal device is a method of managing the data in units of record blocks each consisting of a plurality of sectors.

On the other hand, the scope of use of information processing systems has expanded.

Terminal devices have improved in performance as they have become more complex.

That is, there has been a demand for a practical terminal device that can quickly recover from abnormal processing that occurs during data transfer and that can perform data transfer processing more efficiently.

(Prior Art) As a prior art, a data storage method within a terminal device will be described as an example.

FIG. 2 shows the subsystem within the terminal device. In the terminal device shown in FIG. 2, when writing data to the data file 4 or reading written data, ``Writing/reading to the data file 4 by combining multiple sectors as one record block'' doing.

That is, data input externally or by the own terminal device is temporarily stored in the main memory 2, transferred to the data file 4 under the control of the specified file control channel 3 under the control of the control unit 1, and stored. be done.

The data treated in this way is as described above.

Multiple sectors are collectively stored as one record block in the data file 4.

(1) Improved performance of data file 4 has shortened data transfer time, so data transfer between the processing unit (control unit 1, file control channel 3, etc.) and data file 4 handles multiple sectors at once. It is more advantageous.

(2) Even if data with the same attribute exceeds the capacity of one sector, it is more advantageous for management to handle them all together.

Due to reasons such as.

On the other hand, power supplies for terminal devices generally do not have sufficient recovery measures against power outages due to physical limitations.
Therefore, there is a greater possibility that the power supply will be cut off due to lightning or the like than other devices.

In such a terminal device, if the power is cut off while data is being written to the data file 4, the data update will be interrupted in the middle of a record block, and when the record block is read and confirmed when the power is turned on again, the old and new This is a very inconvenient situation because the data is mixed.

Therefore, conventionally.

(1) A large-capacity capacitor is installed in the power supply to hold the record block for a time equivalent to processing the record block.

(2) Generates an error when reading.

Such methods were used.

[Problem that the invention seeks to solve]

However, if a large capacity capacitor is installed in the power supply, it is necessary to increase the portion occupied by the power supply.

This will increase costs.

Another problem is that if an error occurs during reading when the power is turned on again, it is difficult to determine whether the error is due to the power being turned off or a defect in the data file 4.

[Means for solving problems]

An object of the present invention is to realize a new data file generation management method that solves the above-mentioned problems.

and a control means for updating the code in the generation recognition code area each time data is written to each record block, and the control means updates the code in the generation recognition code area each time data is written in each record block, and the control means updates the code in the generation recognition code area when there is a generation difference between the plurality of generation recognition codes stored in each generation recognition code area. In this case, the problem can be solved by the data file generation management method according to the present invention, which rewrites data to the record block.

[Effect]

That is, 9. For example, generation recognition code areas are provided at the beginning and end of a record block, and a register for storing the generation recognition code is provided in the file control channel. When restarting, the record block is read, the generation recognition code at that time is stored in the register, and when the generation recognition code is examined, 9 For example, the first generation recognition code has been updated, but the last generation recognition code is old data. If it remains the same, it is immediately confirmed that the data is being updated, so a new data transfer request can be made quickly.

〔Example〕

The gist of the present invention will be specifically explained below with reference to an embodiment shown in FIG.

FIG. 1(A) shows an embodiment of a record block configuration according to the present invention, and FIG. 1(B) shows an embodiment of a file control channel within a terminal device according to the present invention. The same symbols indicate the same objects throughout the figures.

Record block a in this embodiment is record block a
Generation recognition code area C at the beginning of the data recording area C
, a generation recognition code area d at the tail end of record block a.

Note that the data in this embodiment is processed in the fourth record block ati, which is composed of a plurality of sectors.

Assume that data having the record block a format shown in FIG. 1(A) is written to the data file 4.

Next, the operation of this embodiment will be explained focusing on the operation of the file control channel 3 shown in FIG. 1(B).

When data from main memory 2 is output to path e,
The bus interface control unit 5 of the file control channel 3 designated by the control unit 1 takes in data and temporarily stores it in a memory (RAM) 9 under the control of a microprocessor 6 (hereinafter referred to as MPU 6).

Next, according to the control program stored in the memory (ROM) 7, the file interface control unit 10 starts the data file 4 of the thio, and transfers the data in the memory (RAM) 9 to the data file 4 according to instructions from the MPIJ 6. , writes and updates transfer data including the record block a format shown in FIG. 1(A) in the corresponding sector (not shown).

If a power failure occurs during this writing, the control unit 1 immediately notifies the file control channel unit 3.

Interrupts writing and updating of transfer data.

゛ Next, when the power is turned on again, the control unit 1 causes the file control channel unit 3 to control the reading of the corresponding record block a of the data file 4, and reads the codes of generation recognition code areas s and d in the read data. The data is stored in the register 8, and the MPU 6 compares and examines the code data stored in the register 8 and detects that the code data in the first generation recognition code areas d and d are different.

The MPU 6 detects the difference between the two code data.

It is determined that the data of the corresponding record block a is in the process of being updated, and the control unit 1 initiates a new data transfer from the main memory 2.
request and rewrite the transfer data.

〔Effect of the invention〕

According to the present invention as described above, even if there is a power outage when writing data to a data file, recovery is possible when the power is turned on again.
This has the effect that it can be processed in a short amount of time and without errors.

[Brief explanation of drawings]

FIG. 1(A) shows an example of a record block configuration according to the present invention. FIG. 1(B) shows an embodiment of a file control channel within a terminal device according to the present invention. Figure 2 shows the subsystem within the terminal device. are shown respectively. In fig. 1 is the control unit, 2 is the main memory. 3 is a file control channel. 4 is a data file. 5Ba bus interface control section. 6 is MPIJ, 7 is memory (R
OM). 8 is a register, 9 is memory (RAM)
. lO is a file interface control unit. are shown respectively. Toru 1 Diagram (A) (β) (Dehetaquil 8) Kaya 2 Prisoner

Claims (1)

[Claims] This is a method for managing data stored in a data file in units of record blocks consisting of multiple sectors, in which generation recognition code areas are provided at multiple locations in the record block, and each time data is written to each record block. and a control means for updating the code in the generation recognition code area, and when there is a generation difference between the plurality of generation recognition codes stored in each generation recognition code area, data rewriting for the record block is performed. A data file generation management method characterized by: