JPH04354036A - Data base management method - Google Patents

Abstract

PURPOSE:To attain the data base management of time sequential data by having an adding position pointer, and a taking-out position pointer controlled by a First In First Out system in a data management. CONSTITUTION:At the time of adding one record, data are filled in fields c1n-c3n in sequence according to the position of an adding position pointer P1 in order to prepare a record rn. Then, one record is written in an invert file corresponding to an object field of a key retrieving processing. Then, the adding position pointer P1 is updated to the next record position rn+1. At that time, when the capacity of a master record MR is filled up, the oldest one record is erased. When the above mentioned capacity has a room to spare, a data adding operation is repeated. At the time of taking out the data, the presence or absence of the data in the master record MR is checked, and one record at the position designated by a present taking-out position pointer P2 is taken out. Then, the value of this taking-out position pointer P2 is updated.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database management method for managing a database set up in an industrial computer system, and more specifically, to an improved method for handling time-series data.

0002

2. Description of the Related Art Conventionally, a database in a computer system installed in a factory is expressed in a two-dimensional table format as shown in FIG. , r2, ... were generally managed. That is, the database creator creates record r1 by filling in fields c11, c21, and c31, and fills in fields c21, c22, and c31.
The master record MR is set by repeating the procedure of filling in 32 and creating record r2.

[0003] Furthermore, such a database (master
There are cases where it is desired to extract only necessary data (records) from records MR), and in this case, data search processing is often performed. Data search processing is often performed by creating an invert file IF near this master record MR. That is, the invert file IF contains a column to be searched later, for example column c.
Regarding 2, this is a file in which the search value "XX" is associated with the record number (r3, r5,...) having this value "XX", and this invert file IF also adds a new record. will be added each time. Therefore, when a key search for the value "XX" in field c2 actually starts, the invert file IF is referenced,
The contents of records r3, r5 are read out from master record MR corresponding to record numbers r3, r5, . . . A plurality of types of invert file IFs are generated in advance in response to all possible search processes. In this way, a database is generated, and a search process can be performed on this database at high speed.

0004

[Problems to be Solved by the Invention] In the conventional database management method as described above, there is no particular chronological registration order for registered data, and the chronological data required by industrial computer systems, etc. It was not suitable for handling. In other words, only key searches are possible for registered data, and in order to perform trend display processing, complicated processing procedures such as setting up a separate program for trend display processing on the application program side are required. It was necessary. The present invention aims to enable time-series data to be managed in a database, and to improve the performance of an industrial computer system.

[0005]

[Means for Solving the Problems] The present invention, which has solved the above problems, creates an invert file that is searched for each key field in a database that has at least one record consisting of at least one field. In a database management method in which search processing is performed using an inverted file generated in advance, a) 1
When adding a record, if it does not exceed the total capacity of the database, add the record; if it exceeds the total capacity of the database, delete the oldest record and then add the record. This database management method is characterized in that it follows an addition position pointer that specifies an addition position, and b) when retrieving one record, it follows a retrieval position pointer that specifies the oldest record in this database.

[0006]

[Operation] When constructing a database using the database management method of the present invention, records are sequentially added to the addition position pointer that specifies the record position by filling in the fields. When retrieving records, if the database capacity is not full, a retrieval position pointer is set to the oldest record, and records are retrieved according to this retrieval pointer. If the database is full, the oldest records are discarded. Refer to the invert file when performing data searches.

[0007]

[Embodiment] An example of implementing the database management method of the present invention will be described below. The database management method of the present invention includes:
In addition to the conventional database management method that allows key searches, FIFO (First In FiFO) is used for data management.
an additional position pointer controlled in the rst Out) manner;
This is characterized by the addition of an extraction position pointer. FIG. 3 is a conceptual block diagram showing the minimum necessary components of a computer system that implements the method of the present invention. The master record MR is a database similar to that shown in FIG. 7, and is a file on the main memory or auxiliary memory of the computer system. The invert file control device IFC is a block that controls the invert file IF shown in FIG. 7, and is located at the same location as the master record MR. The current pointer control device PC is
It is located in a register on the CPU in the computer system or in a fixed area on the main memory, and is involved in position control of the addition position pointer P1 used when adding records and the extraction position pointer P2 used when taking out records. Blocks are also backed up on auxiliary storage if necessary.

The method of the present invention realized in such a configuration is shown in the flowcharts of FIGS. 1 and 2, and its operation will be explained with reference to the conceptual block diagram of FIG. 3. FIG. 2 is a flowchart for constructing a database, and represents the procedure for adding data to the master record MR. When adding one record, data is filled in fields c1n, c2n, and c3n in this order according to the position of addition position pointer P1, and record rn is generated. Then, one record is written to the invert file IF corresponding to the target field of the key search process. For example, if field c2 is assumed to be the key search target, record number rn is registered in the invert file IF corresponding to the value "YY" of field c2n of this record rn. Next, the additional position pointer P1 is set to the next record position (
rn+1). Here, if the capacity of this master record MR becomes full, the oldest record (record r1) will be erased. If there is still room in the capacity of the master record MR, the above data addition operation is repeated.

The flowchart in FIG. 1 is a flowchart for processing for retrieving certain data (records) from this database and for key search processing, while adding data according to the flowchart in FIG. When searching for data, as in the conventional method, the invert file IF to be searched is read out, and then the corresponding record is read out from the master data MR and passed to the application program. When data is to be retrieved, the presence or absence of data in the master record MR is checked, and one record (r1) located at the position specified by the current retrieval position pointer P2 is retrieved. Then, the value of this extraction position pointer P2 is updated. Specifically, in FIG. 3, the extraction position pointer P2 points to this master record M.
Although the oldest record r1 in R is specified, after record r1 is retrieved, the retrieval position pointer P2 moves to the next oldest record r2 after record r1. Next, this read record r1 is delivered to the application program side. Note that when data is extracted, if the data is not in the master record MR, the system waits for data input.

Next, an example of actual operation of data addition, extraction, and retrieval by the method of the present invention will be explained using FIGS. 4, 5, and 6. FIG. 4 is a diagram showing the case where data is added to the master record MR. In this example, the data is initially empty, and the additional position pointer P1 sequentially moves to specify record positions r1, r2, r3, and r4 each time data (records) are added. . On the other hand, during the data addition operation, the retrieval position pointer P2 specifies the oldest data r1 in this master record MR.

FIG. 5 is a diagram showing the case where data is extracted from the state shown in FIG. Initially, the retrieval position pointer P2 specifies the oldest record r1, and when this record r1 is retrieved, it moves to point to the next oldest record r2. In this figure, the retrieval position pointer P2 sequentially moves downward following the successive addition of data, and when it reaches the bottom, moves to the top again. Similarly, in this figure, the additional position pointer P1 sequentially moves downward each time data is added, and when it reaches the bottom, it moves to the top again.

FIG. 6 is a diagram showing a case where data is searched from the state shown in FIG. 4. In field c2, the value "
Assume that you want to extract records that have "ZZ". In this case, as in the conventional method, the corresponding invert
A file corresponding to the value "ZZ" in the file IF is referred to, and a record r3 in which the value of field c2 is "ZZ" is found from this file, and record r3 is read from the master record MR. This executes the data search. Note that this data search process is performed using the additional position pointer P.
1. It is completely unrelated to the position of the extraction position pointer P2.

In this way, in addition to the conventional database management system that allows key searches, FI
It is possible to add an additional position pointer and an extraction position pointer that are controlled by the FO (First In First Out) method.

[0014]

[Effects of the Invention] As described above, according to the database management method of the present invention, the following effects can be obtained. In industrial computer systems, application programs no longer need to perform time-series data management for databases, making it easier for application programs to handle databases. At the same time, searches for specific data required in regular databases can be performed at high speed, and can be applied to a wide range of businesses.

[Brief explanation of the drawing]

FIG. 1 is a flowchart of data search and retrieval in the database management method of the present invention.

FIG. 2 is a flowchart of data addition in the database management method of the present invention.

FIG. 3 is a conceptual block diagram showing components that implement the database management method of the present invention.

FIG. 4 is a diagram showing the state of a master record when a record adding operation is actually performed in the method of the present invention.

FIG. 5 is a diagram showing the state of a master record when a record retrieval operation is actually performed in the method of the present invention.

FIG. 6 is a diagram showing the state of a master record when a record search operation in the method of the present invention is actually performed.

FIG. 7 is a diagram showing conventional general database management. MR Master record IFC Invert file control device PC Current pointer control device

Claims (1)

[Claims] Claim 1: Database management that generates in advance an invert file to be searched for each key field for a database that has at least one record consisting of at least one field, and performs search processing using this invert file. In the method, a) When adding one record, if it does not exceed the total capacity of the database, add the record, and if it exceeds the total capacity of the database, delete the oldest record and then add the record. b) when retrieving one record, following an retrieval position pointer that designates the oldest record in the database.