JPH0764835A - Data storage system for relational data base - Google Patents

Abstract

PURPOSE:To realize a data storage system which increases the access speed of a relational data base and doesn't oppress the capacity of a storage device. CONSTITUTION:When data of the relational data base will be stored in the storage device, data is not stored with a record as unit but is stored with each of data constituting a record as unit, and the address (nf address) indicating the storage position of data of the next record of the same item and the address (nd address) indicating the storage position of data of the next item of the same record are stored together with data. At the time of retrieval, data is compared with a retrieval condition with data like a name as unit, and, when comparison results in coincidence, data of the address, the age, etc., is obtained from the nd address attached to this data. If it doesn't result in coincidence, data of the same item of the next record is obtained from the nf address and is compared with the retrieval condition again.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database data storage system, and more particularly to a relational database data storage system and a data search system.

[0002]

2. Description of the Related Art Conventionally, as shown in FIG. 2, the data storage method of a database is storage in record units.
Then, in order to shorten the data retrieval time, a method of creating an index as shown in the figure is adopted.

Further, Japanese Patent Laid-Open No. 2-181873 discloses a method in which the storage location of a data entity is stored in the same record as the data entity name, so that the data entity can be accessed at high speed from the data entity name. Is listed.

[0004]

However, in the conventional data storage method, when there are a large number of items that can be searched for in a table, it is necessary to create an index for each item, which is extremely extreme. For example, if all the items in the table are search items, it is necessary to create indexes for all the items, which causes a problem of reducing the capacity of the storage device.

When the capacity of the storage device is insufficient, the index creation for some items is abandoned. In such a case, however, the search for the items for which no index is created results in the reading of all records, so the access to the storage device is performed. However, there is a problem in that the efficiency of search processing is reduced as a result.

When the contents of an item for which an index is being created are to be changed or a record is to be added, two accesses occur: access to the data part and access to the index part. There is a problem that the processing time becomes slower than the table without.

[0007]

According to the present invention, as means for solving the above-mentioned problems, in a data storage system of a relational database, each item data constituting a record stores at least the next item data of the same record. It has a data storage system of a relational database characterized by storing in a data unit having an address indicating a location and an address indicating a storage location of the same item data of the next record.

Further, at the time of data search, the item data of the data unit is compared with the search condition, and if different, the same item data of the next record is obtained by the address indicating the storage location of the same item data of the next record. Data storage of relational database characterized by repeating comparison with search conditions and, if they match, repeatedly obtaining the next item data of the same record by the address indicating the storage location of the next item data of the same record The method is used as a means.

[0009]

Embodiments of the present invention will be described with reference to the drawings. In storing the table of the relational database having the values shown in FIG.
Save in a format like.

As shown in FIG. 1, in the data storage system of the present invention, each item data such as customer number, name, address, and age is not stored together in a record unit as in the conventional case, but It is characterized in that the same item next record address (to be referred to as nr address) and the same record next item address (to be referred to as nd address) are attached to each item data and stored in each item data unit. .

Next, a method of retrieving each piece of data stored by the method of the present invention will be described.

For example, when the database management system receives the search command "extract all items of the record registered as Kubota in the name item", the schema information is first read and the top data of the name item is stored. To get the address.

Next, using this address, the data of the first name item is read to determine whether or not the condition is met.

When the condition is not met, the address (nr address) where the data of the same item next record is stored is used to read the name data of the same item next record, and the condition judgment is repeated.

When data that meets the conditions appears, the address (nd) where the data of the next item in the record is stored
Address) is used to sequentially obtain the address, age, etc. as data for one record.

By the series of operations described above, the above-mentioned search can be carried out without creating an index part which has been conventionally required.

FIG. 4 is a schematic flow chart for explaining the search method of the present invention described above.

[0018]

As described above, according to the present invention, data is separately stored for each item, and each data is accessed even when reading or writing, so it is necessary to create an index. In addition, it has an effect of improving the access speed and an effect of facilitating the change of data.

Since it is not necessary to create an index, the capacity of the storage device will not be pressed.

[Brief description of drawings]

FIG. 1 is a diagram showing a data storage structure in a storage device according to the present invention.

FIG. 2 is a diagram for explaining a conventional data storage method.

FIG. 3 is a diagram showing how the data referred to in FIGS. 1 and 2 is recognized on a database.

FIG. 4 is a schematic flowchart for explaining a data search method of the present invention.

Claims (2)

[Claims] 1. In a data storage system of a relational database, each item data constituting a record,
A data storage method for a relational database, characterized in that data is stored in a data unit having at least an address indicating a storage location of the next item data of the same record and an address indicating a storage location of the same item data of the next record. 2. When retrieving data, the item data of the data unit is compared with a search condition, and if the condition is different, the same item data of the next record is determined by an address indicating the storage location of the same item data of the next record. And repeatedly comparing the search condition with the search condition, and if the condition is matched, it is repeated to obtain the next item data of the same record by the address indicating the storage location of the next item data of the same record. The data storage method of the relational database according to claim 1.