JPH05241928A - Data base managing method - Google Patents

Abstract

PURPOSE:To accelerate access by constituting an index file by pairing data in an index field and record numbers of these data record as key data. CONSTITUTION:An index file IF pairs data (such as 'ABCD' or 'XYZU') in a field (f) for respective data records in a master record MR and the physical record numbers (such as 1,2,3... n-1,n) of respective data records r1, r2, r3... rn-1 and rn as one key data Kd (such as 'ABCD1', 'ABCD3'... 'ABCDn-1', 'XYZU2' or 'XYZUn') and the set of this key data Kd is defined as the index file IF. In this case, the respective key data Kd are uniquely correspondent to respective data records. In the case of accessing the index file IF, the access is performed according to this key data Kd.

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 in an industrial computer system, and more specifically to improving the database so that it can be accessed at high speed.

[0002]

2. Description of the Related Art Conventionally, a database in a computer system installed in a factory is, for example, Japanese Patent Application No. 3-129766 "Database Management Method" by the present applicant.
Etc. can be mentioned. That is, generally, as shown in FIG. 6, the database stores each data record r
It is managed as a master record MR composed of 1, r2, ..., rn. Then, there is a case where only necessary data (several data records) is extracted from this database (master record MR) and updated, deleted or added. In such a case, data access is often performed. Done. For data access, an index file IF is often generated near the master record MR and used. That is, the index file IF is
For the field f to be searched later, the data in the field f, for example, "AA01" is combined with the pointer corresponding to the data record r1 having this key data, and further the data "AA02" and this key
It is generated in advance by combining with a pointer corresponding to a data record rn having data, and when executing data access, the index file IF is referred to perform necessary data extraction. There is. A plurality of types of such index files IF are usually generated in correspondence with other index fields.

[0003]

The conventional database management method as described above is effective in that the search processing by key data can be performed at high speed by utilizing the index file. When updating or deleting a file, the same key
If there are multiple data records containing data, it is inefficient and time consuming because all the pointers corresponding to this same key data must be searched to update the index file. Was there. The present invention solves such a problem, and an object thereof is to realize simple and high-speed database management.

[0004]

The present invention which achieves the above object is a key in the data record for a database having at least one data record including at least one field. In a database management method in which an index file is generated in advance for each field and a search process is performed using this index file, when accessing the index file, a key that combines the data in the field to be the index and the data record number It is a database management method characterized by accessing by data.

[0005]

According to the database management method of the present invention, when the index file is accessed, the key data obtained by synthesizing the data in the field serving as the index and the data record number is accessed.

[0006]

The present invention will be described in detail below with reference to the drawings. FIG. 1 is a diagram showing the concept of the database management method of the present invention. In this figure, data records r1, r2
The master record MR composed of r3, ..., rn-1, rn is a database equivalent to the conventional one shown in FIG. 6, and the structure of the index file IF is characteristic of the present invention. That is, the index file IF is the master record MR
The data ("ABCD", "XYZU", etc.) in the field f of each data record in the data record and the physical record number of each data record r1, r2, r3, ..., rn-1, rn (this In the example, 1, 2, 3, ..., N-
, N) and one key data kd ("A
BCD 1 "," ABCD 3 ", ...," ABCD n
-1 ", ...," XYZU 2 "," XYZU n ", etc.)
And the aggregate of this key data kd is the index file I
Let it be F. Each key data kd in this index file IF uniquely corresponds to each data record. The management block B in the master record MR is
It is a block that manages each data record and has no direct relationship with the present invention.

The operation concept of the present invention will be described with reference to the flow charts of FIGS. 2, 3, 4, and 5. FIG. 2 shows an operation when a new data record is added to the master record MR. In this case, there is nothing new in comparison with the conventional operation, and when data is written in all the fields of the data record to be added in the master record MR, this The key data is combined by combining the field data and the physical number of the data record, and the index file IF
The index file IF is generated by being sorted together with other key data in the index file IF. Then, in this newly added data record, there may be another index field as an index, and in this case, this index file is also updated. In this way, when a new data record is added, the index file IF is also rewritten.

FIG. 3 shows an operation when a certain data record in the master record MR is deleted.
When the data serving as the index is deleted during the data record deletion, the data serving as the index and the data record number to be deleted are known in advance, and the key data kd is composed from the data record number.
To search the index file IF. When a key data record is found in this index file, it is deleted. Subsequently, key data is similarly generated for the other index fields in this data record, and the index file is searched to delete the found record. Finally, the data record in the master record MR is deleted.
Thus, in data record deletion, the record of the corresponding key data in the index file IF can be uniquely deleted.

FIG. 4 shows a data search operation. In this case, there is no great difference from the conventional operation. If the data to be the key of the search is given and the field storing this data is the index field, the index file IF is searched using this as a key and the corresponding data
Get the record If the key field is not the index field, all data records in the master record MR are searched to obtain the corresponding data record.

FIG. 5 shows an operation when updating a certain data record in the master record MR.
At the time of updating the data record, the index file IF is searched by the key data kd which is synthesized from the data to be the index before the update and the data record number to be updated, and the data record number. When a key data record is found in this index file, it is deleted. Then, the updated data and the data record number are combined and added to the index file after sorting. Then update the actual master record. Next, for the other index fields in this data record, the key data before update is similarly generated, the index file is searched to delete the found record, and the data is updated. In this way, in updating the data record, the record of the corresponding key data in the index file IF can be uniquely updated and deleted.

As described above, according to the method of the present invention, the index file is constructed as key data by combining the data in the index field and the record number of the data record. , The correspondence between the index file IF and the data record in the master record MR becomes unique, and further, the index file IF
Are sorted, the number of accesses to the index file IF is reduced in data update and deletion, and the access speed is increased as a whole.

[0012]

As described above, according to the present invention,
Even if the key data of the index file and the data record in the master record uniquely correspond to each other and the data of the field that serves as the index is duplicated, the index file can be changed quickly and efficient database management is possible. You can get the way.

[Brief description of drawings]

FIG. 1 is a diagram showing the concept of the method of the present invention.

FIG. 2 is a diagram showing an operation example of the method of the present invention.

FIG. 3 is a diagram showing an operation example of the method of the present invention.

FIG. 4 is a diagram showing an operation example of the method of the present invention.

FIG. 5 is a diagram showing an operation example of the method of the present invention.

FIG. 6 is a diagram showing a conventional method.

[Explanation of symbols]

MR master record IF index file kd key data r1, r2, r3, ..., rn-1, rn data record

Claims (1)

[Claims] 1. A database having at least one data record including at least one field, an index file is generated in advance for each key field in the data record, and a search is performed by this index file. In the database management method for processing, when the index file is accessed, access is performed by key data obtained by synthesizing data in a field serving as an index and the data record number.