JPH02297640A - Data access system for distributed data base system - Google Patents

Abstract

PURPOSE:To shorten the time for processing by executing write processing where a file for which data processing is already executed is written to a data base by time sharing processing using the idle time during data processing operation to the next file. CONSTITUTION:When the data processing operation such as addition, change and deletion, etc., to the data of a file 6 is executed and finished, a work station WS 2 does not immediately start the write processing to a data base 3 to this file 6, for which the data processing is already executed, but accepts read operation to a next data file 7. When the read operation to the next file 7 is executed, the read processing to the file 7 is executed and the data processing operation of the file is accepted. When an operator starts the data processing operation to the file, the data processing operation is accepted and write processing to a data base 3 in the preceding file 6, for which the data processing is already executed, is started. Thus, since the write processing is executed by utilizing the idle time during the data processing operation, processing speed is improved as a whole.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention relates to a distributed database in which a plurality of workstations are connected via a transmission line and data is distributed and stored in the database of each workstation. The present invention relates to systems, and particularly to a data access method for a distributed database system that efficiently accesses arbitrary files stored in each database.

(Prior art) In a LAN (local area network) in which multiple workstations (hereinafter abbreviated as WS) are connected via transmission lines, a database is distributed and embedded in each WS without using a host computer. There is a distributed database system.

FIG. 6 is a diagram showing a system in which, for example, various information on each patient in a hospital is stored in a database, and various data on each patient can be searched and updated from each WS. That is, each wS21:l: data base 3.1 connected to the transmission path 1. A CRT display section 4, a keyboard 5, etc. are incorporated.

In the database 3 of each WS 2, as shown in FIG. 7, an order file 6 and an implementation input file 7 are created for each patient. And since the number of patients is large,
It is stored across the databases 3 of multiple WS2.

The order file 6 stores various order data that the doctor instructs the nurse for the patient. Further, the implementation input file 7 stores various implementation data for the patient, which the nurse performed based on the doctor's instructions. Therefore, the order file 6 is searched by doctors and nurses, whereas the implementation input file 7 is searched mainly by nurses.

FIG. 8 is a flowchart showing the processing procedure when the doctor executes data processing operations such as adding, changing, and deleting data in the order file 6 for a given patient, and FIG. It is a figure which shows the update and the update procedure of the implementation input file 7 by a nurse.

First, when a doctor executes a file read operation on the order file 6 of the patient with patient number i in one WS2 using, for example, date data as a search key, the order file 6 of the corresponding patient number i is stored in the database 3 of each WS2.
The information is searched from within, read out to the data buffer of the WS2 operated by the doctor, and displayed on the CR1 display section 4.

Then, the doctor uses the keyboard 5 to perform data processing operations such as addition, modification, and deletion on each data of the i-th order file 6 read out to the CR1 display section 4. When the predetermined data processing operation is completed, the doctor executes a write operation to write the data-processed first order file 6 to the original database 3. Specifically, a write command is input using the keyboard 5. The WS 2 to which the write command has been input writes the i-th order file 6 to the original database 3.

The above update operation for the order file 6 is performed sequentially for each patient's order file 6.

Next, the nurse, for example, in one WS 2 different from that of the doctor, executes a file read operation for the order file 6 of the patient with patient number i, using the date data as a search key, as described above.

Then, the order file 6 of the corresponding patient number i is stored in each WS2.
is searched from within the database 3 of
will be displayed.

The nurse performs a read operation for the i-th implementation human input file 7 from the database 3 using a similar operation, reads this i-th implementation input file 7 to her own WS 2, and displays it on the CR1 display unit 4. .

Then, the nurse looks at each data specified by the doctor in the order file 6, and based on that data performs a data processing operation to store (copy) the added, changed, and deleted data as new items in the input file 7. is executed, and with respect to the implementation input file 7 to which the new item has been added, implementation result data corresponding to the doctor's instructions is written into the implementation manual file 7 using the keyboard 5. That is, the data processing operation on the implementation input file 7 is executed.

When the predetermined data processing operation is completed, the nurse executes a write operation to write the data-processed i-th implementation input file 7 into the original database 3.

Specifically, a write command is input manually using the keyboard 5. The WS 2 to which the write command has been input writes the i-th execution input file 7 to the original database 3.

However, each file 6 stored in each database 3 of the distributed database system is
．． There were still problems with the method of updating data in 7.

That is, when updating the order file 6 and implementation input file 7 for multiple patients in succession, as shown in FIG. Read processing, data processing, and writing processing for the order file 6 are executed in order.

Similarly, when a nurse updates data in the implementation input file 7, the order file 6 is read.

Data processing for reading data from the implementation input file 7, adding new items, data processing for inputting implementation result data, and writing processing are executed in order.

Therefore, whether it is the doctor's side or the nurse's side,
Unless a series of processes for one patient's file 6.7 is completed, processing for the next patient's file 6.7 cannot be started. Therefore, there is a problem in that as the number of patients increases, the time required to update data increases.

On the other hand, each file 6.7 is displayed on the CRT display section 4 by the doctor and the nurse by key operations on the keyboard 5.
In data processing operations such as adding, changing, and deleting data, the processing speed of WS2 is much faster than the key human power speed, so the key human power waiting time increases for WS2.
As a result, the processing capacity of WS2 is not used effectively.

(Problem to be solved by the invention) As described above, according to the conventional data access method, as the number of files to be updated increases, the time required to update the data increases correspondingly to the number of files, and conversely, the free time increases. This results in the inconvenience of a paradoxical phenomenon in which the amount of energy increases.

The present invention was made in view of these circumstances, and
By performing the write process of writing data-processed files to the database during data processing operations for the next file, the time required for the entire process can be shortened when multiple files are updated consecutively. The purpose of this invention is to provide a data access method for a distributed database system that can improve the operational responsiveness of the entire system and greatly improve the efficiency of file update operations.

[Structure of the Invention] (Means for Solving the Problems) In order to solve the above problems, the data access method of the distributed database system of the present invention connects a plurality of workstations each having a database via a transmission line. In a distributed database system in which multiple files storing multiple pieces of data as one unit are distributed and stored in each database, each workstation reads a specified file from the database in response to a file read operation. Performs read processing, performs specified data processing on the data of the read file in response to the data processing operation, and performs write processing to write the data-processed file to the original database before the next file read operation. This is a time-sharing process that utilizes free time during the data processing operation period for the file read out.

(Operation) With the data access method configured in this way, when data is updated for multiple files in succession, when a read operation is performed on one file, the corresponding file is read from the database. . Then, data processing operations such as addition, modification, and deletion are performed on the data in this file. When this data processing operation is completed, the workstation (WS) accepts a read operation for the next data file without immediately starting writing processing to the database for this data processed file.

Then, when a read operation is performed on the next file,
Execute read processing for the file and accept data processing operations for the file. When the operator starts a data processing operation for a corresponding file, the data processing operation is received 44 and writing processing of the previously data-processed file to the database is started. In other words, since this write processing is performed using the free time of data processing operations, the time required for data processing in response to data processing operations does not increase, so conventional write processing can be performed independently. The overall processing speed will be increased compared to when it was implemented.

(Example) An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of a distributed data system employing the data access method of the embodiment. The same parts as in FIGS. 6 and 7 are given the same reference numerals, and the explanation of the overlapping parts will be omitted.

That is, each WS2 connected to the transmission line 1 has a database 3. CR7 display section 4. A keyboard 5 etc. is incorporated. In the database 3 of each WS 2, as shown in FIG. 7, an order file 6 and an implementation input file 7 are created for each patient. Since the number of patients is large, each file 6.7 has multiple WS2 databases 3.
It is stored over the period.

Furthermore, each WS 2 is provided with a data buffer for temporarily storing each file 6.7 read from each database 3 and data-processed files, and a file management information memory 8 for storing a file write reservation list, etc. .

FIG. 2 is a flowchart (main routine) showing the operation of the WS 2 when a doctor executes data processing such as adding, changing, or deleting data in an order file 6 for multiple patients. Processed file 6.
7 is a flowchart (subroutine) showing a write process for writing 7 to the original database 3. FIG. 4 is a diagram showing the procedure for updating the order file 6 by the doctor and the procedure for updating the implementation manual file 7 by the nurse.

In the main routine of FIG. 2, in S (step) 1, when a file reading operation for the order file 6 of the patient with patient number i is executed by the doctor's key operation,
At S2, execution of the write processing subroutine shown in FIG. 3 is temporarily stopped.

Next, in S3, the order file 6 of the corresponding patient number l is retrieved from the database 3 of each WS2 and read into the data buffer of the WS2. It is displayed on the CR7 display section 4 at the same time.

When the reading process of the order file 6 is completed, in S4,
The write processing subroutine of FIG. 3 for the previous data-processed order file stored in the data buffer is activated. Then, additions are made to the order file 6 from the keyboard 5.

Waiting for data processing operation input such as modification or deletion.

When a data processing operation command is input from the keyboard 5, data processing such as addition, modification, deletion, etc. is performed on each data in the corresponding data file 6 in S5.

When the data processing end operation is performed on the keyboard 5,
In S6, the order file 6 of the patient number i whose data has been processed is stored in the data buffer, and the number i of the corresponding order file 6 is registered in the file write reservation list.

When the file reading operation for the next order file 6 of patient number j is performed in S7, the process returns to S2, the subroutine for file writing processing is temporarily stopped, and the reading process of the order file 6 of patient number j is performed in S3. Execute.

If the file reading operation for the order file 6 of the next patient number is not executed in S7, this main routine is ended.

When the write processing subroutine is executed at step 84 in FIG. 2, the process proceeds to FIG. , reads the corresponding number from the file write reservation list, and starts writing processing to the database 3 for the order file 6 with the corresponding number stored in the data buffer. When the writing process for one order file is completed, the process returns to 88, and it is checked whether or not the number of the order file 6 that has not been written yet exists in the file write reservation list, and if it does not exist, this write Terminates the input processing subroutine.

Note that this write processing subroutine is executed in a time-sharing manner using the free time of each processing in the main routine.

That is, the write processing subroutine in Figure 3 starts when the data processing operation in the main routine in Figure 2 is started, and temporarily suspends the write processing operation when a read operation for a new order file is executed. do. That is,
The data processed order file 6 is executed during the data processing operation for the next order file 6.

Note that the processing time required for writing the order file 6 into the database 3 is a time that can be sufficiently processed within the waiting time of the key operation during which the doctor is performing the data processing operation. Therefore, the writing process of the data file 6 to the database 3 is almost completed during the data processing operation period for the next data file.

The nurse reads each patient's order file 6 and implementation personnel file 7, executes data processing operations on the implementation input file 7, and creates a data-processed implementation input file 7.
Even when writing to the original database 3, each WS
The writing process of the data-processed implementation input file 7 in step 2 is executed in a time-sharing process using free time during the data processing operation for the next implementation input file 7, as described above.

Figure 5 shows the operations performed by doctors and nurses in each WS2, each process 1 performed by the corresponding WS2 in response to the operations, and the operations performed by each WS2 independently when such a data access method is adopted. 3 is a time chart showing the implementation timing of time-sharing processing to be performed.

That is, the doctor reads the i-th order file 6 and performs a data processing operation on the corresponding order file 6. Next, the j-th order file 6 is read and data processing operations are performed on the corresponding order file 6.

In this case, the i-th data processed order file 6 is j
is written to the original database 3 during the data processing operation for the order file 6.

On the other hand, the nurse reads the i-th order file 6, then reads the i-th implementation input file 7, performs data processing to add new items to the read i-th implementation input file 7, and then performs the same process. A data processing operation for inputting implementation result data to the i-th implementation input file 7 is executed.

Next, read the j-th order file 6 in the same way,
Next, the j-th implementation input file 7 is read, the data processing operation of adding new items is executed to the read i-th implementation input file 7, and then the implementation result data input to the same i-th implementation input file 7 is performed. Perform data processing operations.

In this case, the i-th data-processed implementation manual file 7 is written to the original database 3 during the data processing operation of adding new items to the j-th implementation input file 7. Note that the writing process of the i-th data-processed implementation manual file 7 to the database 3 may be continuously executed even during the data processing operation of inputting implementation result data to the implementation manual file 7.

In this way, each data processed file 6.7 is stored in the database 3 during the data processing operation for each next file.
Therefore, when the data processing operation for each file 6.7 is completed, it becomes possible to immediately perform a read operation for the next file 6.7.

Therefore, as is clear from the comparison with the conventional method shown in Fig. 10, the time interval between the update work of one file and the update work of the next file by an operator such as a doctor or nurse can be shortened. , the work efficiency of file update operations can be greatly improved.

Note that the present invention is not limited to the embodiments described above. In the embodiment, various data of each patient in a hospital is stored in a file, but it goes without saying that the present invention is not limited to hospitals and can be applied to searching and updating various data in various facilities.

[Effects of the Invention] As explained above, according to the access method of the distributed database system of the present invention, the write process of writing a data-processed file to the database is performed by utilizing free time during data processing operations for the next file. This is done using time-sharing processing. The data processing operation described above is generally performed by an operator using, for example, a keyboard, and there is sufficient free time during the data processing operation. Therefore, when updating a plurality of files in succession, the time required for the entire process can be shortened, the operational responsiveness of the entire system can be improved, and the work efficiency of file updating operations can be significantly improved.

[Brief explanation of the drawing]

1 to 5 show a data access method of a distributed database system according to an embodiment of the present invention, and FIG. 1 is a block diagram showing a distributed database system adopting the data access method of the embodiment, FIGS. 2 and 3 are flowcharts showing operations, FIGS. 4 and 5 are diagrams showing operating procedures, FIG. 6 is a block diagram showing a distributed database system adopting a conventional data access method, and FIG. Figure 7 is a diagram showing each file formed in each WS,
FIG. 8 is a flowchart showing the conventional method, and FIGS. 9 and 10 are diagrams showing the operating procedure of the conventional method. 1...Transmission line, 2...Work station (WS)
, 3... Database, 4... CRT display section, 5.
...Keyboard, 6...Order file, 7...Implementation input file, 8...File management memory. Applicant's representative Patent attorney Takehiko Suzue Figure 1 Figure 2 Figure 8 Figure 9

Claims (1)

[Claims] A plurality of workstations each having a database are connected via a transmission line, and a plurality of data are
In a distributed database system in which a plurality of files stored as a unit are distributed and stored in each of the databases, each workstation performs data processing by reading a specified file from the database in response to a file read operation. The specified data processing is performed on the data of the file read in response to the operation, and the write process of writing the data-processed file to the original database is performed on the file read in the next file read operation. A data access method for a distributed database system characterized by performing time-sharing processing using free time during a data processing operation period.