JPH0619770A - Data update processing method for common file - Google Patents

Abstract

PURPOSE:To efficiently execute an update processing with a simple mechanism without being conscious of the advance state of the processing in the other access source by providing the data management mechanism managing respective processing situations on a common file-side. CONSTITUTION:The data management mechanism 5 setting-alters a common file 2 to be unprocessed, in the middle of update processing and update processed. Respective work stations 1 inquire the data management mechanism 5 of the processing situation whenever one piece of update data 4 is read, and execute the update processing as to update data from which the response of 'unprocessed' is obtained. Update data from which the responses of 'update processing' and 'update processed' are obtained are skipped and ignored. Thus, the operators of the respective work stations 1 can advance the update processing at a self pace without being afraid of executing the duplicate update processing of same data in the common file 2 and without being conscious of the advance state of the update processing in the other work station.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an update processing method in which access sources such as a plurality of workstations jointly update data in a shared file.

In order to quickly update one file for a large amount of input update data, a file is shared by a plurality of workstations, and the update data is divided and distributed to each workstation, and a plurality of operators are provided. There is a processing system in which the update processing can be performed jointly from different workstations. The present invention provides an update processing method capable of efficiently executing update access in which a plurality of access sources jointly execute a large amount of update data for such a shared file in parallel.

[0003]

2. Description of the Related Art FIG. 3 shows a plurality of workstations.
It is a schematic block diagram of the conventional processing system which shares one file. In the figure, 1-1, 1-2, ..., 1-
N is a workstation, 2 is a shared file accessed by each workstation, 3 is a network, and 4-1, 4-2, ..., 4-N are update data.
The shared file 2 is actually managed by one workstation.

The data of the shared file 2 is accessed in record or block units, and the data is # 0, # 1, ..., #.
Identified by M. In addition, here, each data needs to be updated in the input order (occurrence order), and within a certain update period or a group of certain update data,
It is not allowed to update the same data of the shared file 2 multiple times.

The operator of each workstation updates the shared file with respect to the update data given thereto. For example, when the workstation 1-1 issues an update request to the shared file 2 for the update data of # 0, the shared file 2 accepts the update request if the data of # 1 is in the process of being updated or has not been updated. 1 is updated and the result is notified to the workstation 1-1 of the access source. The workstation 1-1 of the access source sends an update request for the next update data to the shared file 2 when the shared file 2 notifies the update end.

If the work ability of the operator of each workstation is different, the progress of the processing of the update data is different between the workstations. The update data distributed to each workstation is a group of update data (for example, 1
Since they belong to a group of update data based on input data for a day), it is desirable that those update processes end almost at the same time and the update data of the next group be processed at the same time. Therefore, each workstation needs to be aware of the progress of processing in other workstations. As a solution to this, it is conceivable to distribute the update data in an amount according to the ability of the operator, but it has a drawback that it is difficult to determine the distribution amount and management becomes troublesome. As another solution, a distribution program for collectively managing distribution of update data is provided, and the distribution program is activated in response to a request for update data from each vacant workstation, and the update data is distributed one by one. However, there are drawbacks that the processing becomes complicated and the amount of required resources increases.

[0007]

SUMMARY OF THE INVENTION According to the present invention, when a plurality of access sources jointly execute update processing of a shared file that requires a guaranteed data update order, the progress of the processing at the other access sources can be confirmed. It is intended to provide a means that can be efficiently executed by a simple mechanism without needing to be aware.

[0008]

According to the present invention, the shared file side is provided with a data management mechanism for managing the processing status of each data during update processing, updated processing, and unprocessed processing. , Update data is skipped except when the update access target data in the shared file is unprocessed, and a read-skipping management mechanism is provided to execute only the update data that can be processed. Inquires the management system about the processing status, and if the update access target data in the shared file is in the process of being updated or has been updated, it is assumed that the order of data update may be destroyed and the update data that was being processed is read. It skips and moves to the processing of the next update data.

FIG. 1 illustrates the principle of the present invention. In the figure, 1 is a workstation which is an access source. Only one of the workstations is shown here for simplicity of illustration.

Reference numeral 2 is a shared file accessed by a plurality of workstations. The data to be accessed is identified by # 0, # 1, ..., #N.

Reference numeral 4 denotes update data to be processed by the workstation 1, which is input using a form. A data management mechanism 5 is provided for managing the processing status of each data of the shared file 2. It also has the function of responding to the inquiry about the data processing status.

Reference numeral 6 is a management file for distinguishably managing the processing status of whether update processing is in progress, update processing has been completed, or unprocessed for the data of the shared file 2. Referenced and updated.

Reference numeral 7 denotes a read skip management mechanism provided in the workstation 1, which skips the update data 4 one by one and inquires of the data management mechanism 5 that the update processing is in progress or the update processing has been completed. If a response is made, the process skips to the next update data, and only if an unprocessed response is made, the update processing by the update data is executed.

[0014]

In FIG. 1, the data management mechanism 5 sets the processing status of the management file 6 to "unprocessed" for all data before the update processing from the workstation (1) is started for the shared file 2. Initialize to the state. Next, when one of the workstations starts the update process for the “unprocessed” data, the processing status of that data is changed to “update process in progress”, and when the update process of that data is completed, Change the status to "Updated". In this way, the content of the management file 6 is updated according to the update of the shared file 2.

Each workstation (1) sequentially reads and processes a plurality of update data 4 distributed to itself. Every time one piece of update data 4 is read, the data management mechanism 5 is inquired about the processing status, and the update processing is executed only for the update data for which an “unprocessed” response has been obtained, and “update processing in progress” and “update processing completed” are executed. The update data for which a response of “” is returned is skipped and ignored.

As a result, the operator of each workstation (1) does not have to worry about performing duplicate update processing on the same data in the shared file 2 and does not need to be aware of the progress of the update processing at other workstations. In addition, the update process can proceed at its own pace.

[0017]

2 is a block diagram of a processing system according to an embodiment of the present invention. In FIG. 2, 1-1, 1-
Reference numerals 2, 1-3, 1-4 are workstations, of which 1-1, 1-2, 1-3 are workstations that jointly execute processing of update data, and 1-4 is a shared file. It is a managing workstation.

Reference numeral 2 is a shared file. 3 is a network connecting the workstations. 4-
1, 4-2 and 4-3 are workstations 1-1 and 1
It is the update data distributed to −2 and 1-3 respectively.

Reference numeral 5 is a software data management mechanism provided in the workstations 1-4. Reference numeral 6 is a management file.

Reference numerals 7-1, 7-2 and 7-3 are software skip control mechanisms provided in the workstations 1-1, 1-2 and 1-3. The workstations 1-1, 1-2, 1-3 independently execute the processing of the update data 4-1, 4-2, 4-3 distributed to themselves. Therefore, each read skip management mechanism 7-1, 7-2,
7-3 inquires of the data management mechanism 5 of the workstation 1-4 for each update data. On the other hand, the data management mechanism 5 uses the management file 6
To notify the processing status displayed in the data. Upon receiving the "unprocessed" notification, the read skip management mechanism of the workstation that made the inquiry notifies the data update processing mechanism (not shown) in the workstation that update processing can be executed for the update data. . As a result, the data update processing mechanism issues an update processing request for the shared file 2 to the workstation 1-4 according to the normal procedure.

Upon receipt of the update processing request, the data management mechanism 5 of the workstation 1-4 changes the display of the corresponding data in the management file 6 to "update processing in progress", and after the update processing is executed, the display is further updated. Change to "Processed".

The data management mechanism 5 that received the inquiry
Is displayed in the corresponding data of the management file 6 "Updating"
Alternatively, when “update processing completed” is found, the fact is notified to the workstation of the access source, and the read skip management mechanism of the notified workstation skips the update data and moves to the next update data.

When the processing of the last update data in each workstation is completed by repeating the above operation, the update data skipped by then remains. Since the remaining update data may destroy the update order of the data of the shared file 2, it is separately processed after the check.

[0024]

According to the present invention, in a plurality of access sources that jointly execute processing of a large amount of update data for a shared file, the update data distributed to itself is used as the progress of processing of other access sources. It is possible to implement an efficient processing system that is easy to manage and that can be processed at its own pace regardless of.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a block diagram of a processing system according to one embodiment of the present invention.

FIG. 3 is a schematic block diagram of a conventional processing system.

[Explanation of symbols]

 1 Workstation of access source 2 Shared file 4 Update data 5 Data management mechanism 6 Management file 7 Skip management mechanism

[Procedure amendment]

[Submission date] July 15, 1992

[Procedure Amendment 1]

[Document name to be corrected] Drawing

[Correction target item name] All drawings

[Correction method] Change

[Correction content]

[Figure 1]

[Fig. 2]

[Figure 3]

Claims (1)

[Claims] 1. A processing system in which a plurality of access sources share one file and each of which can issue a data update request to the shared file, wherein each shared file data is provided on the shared file side. Is provided with a data management mechanism that manages whether the processing status is being updated, has been updated, or has not been processed, and responds to the inquiry with the processing status. For each update source, query the data management mechanism for the processing status for each update source.If a response indicating that the update is in progress or has been updated is returned, the update data is skipped and the next update data is sent. Then, the data management mechanism is equipped with a skip management mechanism that inquires about the processing status, and each of the above multiple access sources When saving the update order according to the update data generation order and performing update processing with multiple update data jointly, the update data management mechanism is queried via its own skip management mechanism and the update processing is in progress. Alternatively, a method for updating data in a shared file is characterized in that the update data for which an updated response has been returned is skipped and the update processing is sequentially executed only for update data for which an unprocessed response has been returned.