JPH09311801A - Data base device and data base retrieval method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the frequent occurrence of insufficient memory capacity and swap-out states and to effectively retrieve data, by reading and storing only the data that is specified by a recognized read pattern. SOLUTION: A secondary storage 3 stores the read patterns of plural data. When the desired one of these read patterns is designated, a control part 4a recognizes the designated pattern and reads only the data that is specified by the recognized pattern out of the storage 3 to store it in a memory part 4b. Thus, only the data that is specified by the read pattern recognized by the part 4a is stored in the part 4b. Then an arithmetic part 4c extracts the data accordant with the retrieval condition out of the data stored in the part 4b. As a result, the data retrieval response is improved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database device for operating a database and a database search method performed using this database device, and more particularly to an object-oriented database device and its database search method.

[0002]

2. Description of the Related Art A database device is a computer system having hardware and software for operating a database, and stores data groups having a systematic data structure and searches the stored data. It is something to do. Conventionally, as a database device, a relational (relational) database device having a table (table) data structure is widely known. However, there is a problem that the relational database device cannot handle all the required jobs and it is difficult to realize high-speed processing. Therefore, in recent years, instead of the relational database device, an object-oriented database device that can perform high-speed processing even with a complicated data structure has been used.

The object-oriented database device is
A database is constructed by using object data, which is a combination of data and procedures related thereto, as one module (unit). Therefore, in the object-oriented database device, whereas the relational database device can store only numerical values and characters,
It can store a variety of data such as graphics and moving images, and can pseudo-support whether the data structure is a relational type, a hierarchical type, or a network type. It can be stored as a database together with logical data.

Further, the object-oriented database device has, as its internal functions, a secondary storage management function, a compound object having a structure such as an array, and an object ID.
It has features such as data association by, and object-specific function support such as "inheritance". The object ID is an identifier (ID) given to each piece of object data, and is used to perform data association between pieces of object data, as shown in FIG. 5, for example.

In such an object-oriented database device, a magnetic disk or MO (Magnet Optic) is usually used.
A storage unit such as an al Disk (magneto-optical disk) and a CPU (Cent
ralProcessing Unit) and RAM (Random Access Memor)
y), etc., and a search processing unit that is a combination, and the search processing unit performs a data search according to the search condition from the object data accumulated in the storage unit to construct the database. .

For example, as shown in FIG. 6, when the storage unit stores object data in a table format including a plurality of records (columns) and a plurality of fields (rows), the search processing unit: Data retrieval is performed according to the procedure shown in the flowchart of FIG. However, in FIG. 6, it is assumed that the table-type object data has a data amount of 38 bytes for one record (inside the thick line in the figure). Further, it is assumed that the memory such as the RAM included in the search processing unit has a data area capacity of 300 bytes.

As shown in FIG. 7, when a user (hereinafter, referred to as a user) of this object-oriented database device specifies search conditions, the search processing unit causes the memory of the search processing unit to store The reading of all the object data accumulated in the storage unit is started (step 201, hereinafter step is abbreviated as S). At this time, the search processing unit reads the object data accumulated in the storage unit for each record in the table format (see FIG. 6). Therefore, if the data amount of one record of object data is 38 bytes and the data capacity of the memory is 300 bytes, the search processing unit will store 300 bytes of object data that can be stored in this memory, that is, 7 records. Object data (38 bytes × 7 = 266 bytes) is read from the storage unit and stored in the memory.

Here, the search processing unit compares the object data of seven records stored in the memory with the specified search condition by the CPU or the like of the search processing unit, and compares the object data with the specified search condition. It is determined whether there is matching object data (S202). As a result of this judgment, if there is object data that matches the search condition,
The search processing unit leaves the object data in the memory and swaps out the object data that does not match the search condition (S203). Swap-out is, for example, as shown in FIG. 8, sending the object data read from the storage unit to the memory of the search processing unit to the database in the storage unit again for storage.

Further, in FIG. 7, when object data that does not match the search condition is swapped out, the search processing unit newly reads the object data of the swapped-out data amount from the storage unit, and again loads it into the memory of the search processing unit. Store. That is, the search processing unit reads the object data of the eighth record and thereafter from the storage unit, and repeats the above processing for the object data (S201 to S203). Then, for all the object data accumulated in the storage unit, when the comparison with the search condition and the residual or swap-out of the object data as a result of this comparison are completed, the search processing unit
The object data stored in the memory is used as the data search result, and the CRT (Cath
ode Ray Tube) and display it on the screen (S204).

As described above, in the object-oriented database device, the object data stored in the storage unit is searched for the object data that matches the specified search condition. At this time, in the object-oriented database device, since the respective object data accumulated in the storage unit are associated with each other by the object ID, all these object data are read from the storage unit into the memory of the search processing unit. Enables navigational search.

[0011]

By the way, in the above-described object-oriented database device, when performing a data search, all the object data accumulated in the storage section is temporarily read into the memory of the search processing section. Has become. Therefore, if a large amount of object data is accumulated in the storage unit, there is a high possibility that the memory capacity of the search processing unit will become insufficient, and as a result, swap-out will occur frequently. This is the same even if, for example, it is desired to retrieve only a few object data stored in the storage unit.

That is, in this object-oriented database device, even if only a few object data are required, for example, all the object data in the storage unit are read into the memory once.
Insufficient memory capacity and frequent swap-outs occur, and as a result, the efficiency (response) of data retrieval decreases.

In view of the above problems, it is an object of the present invention to provide a database device capable of suppressing insufficient memory capacity and frequent occurrence of swap-out, and efficiently performing data retrieval. . Further, in view of the above problems, it is another object of the present invention to provide a database search method capable of suppressing insufficient memory capacity and frequent occurrence of swap-out, and efficiently performing data search.

[0014]

DISCLOSURE OF THE INVENTION The present invention is a database device devised to achieve the above object, and stores data storage means for storing a plurality of data and data read from the data storage means. A storage means and a search means for extracting data matching the search condition from the data stored in the storage means, further comprising presetting among a plurality of data stored in the data storage means. A pattern storage means for storing a set of read data as a read pattern, and a desired read pattern is designated from the read patterns stored in the pattern storage means prior to the extraction of the data by the search means. , When the recognition pattern that recognizes this and the recognition pattern is recognized by this recognition mechanism, In which only the data specified comprising a reading means for storing in said storage means reads from said data storage means I.

According to the database device having the above structure, the pattern storage means stores the read patterns for the plurality of data in the data storage means. Here, when a desired read pattern is specified from the read patterns, the recognition means recognizes this, and the read means reads only the data specified by the read pattern from the data storage means and stores it. Store in the means. As a result, only the data specified by the read pattern recognized by the recognition means is stored in the storage means, and the search means extracts the data satisfying the search condition from the data in the storage means. In other words, in this database device, if the reading pattern corresponding to the search condition for the search means to extract the data is specified, not all the data in the data accumulating means, but the data specified by the reading pattern, Data that meets the search conditions is extracted.

Further, the present invention is a database search method devised to achieve the above object, and data storage means for storing a plurality of data, and storage means for storing the data read from the data storage means. A method of using a database device having a search means for extracting data satisfying a search condition from the data stored in the storage means, the method including: a plurality of data stored in the data storage means; A process of setting a set of arbitrary data as a read pattern, a process of storing the set read pattern together with the plurality of data in the data accumulating unit, and a read pattern stored in the data accumulating unit. Only the process that specifies the read pattern and the data specified by the specified read pattern Wherein a process of storing in said storage means reads from the data storage means, characterized by comprising the said storage means and said determining whether match the search with respect to the data stored in the processing.

According to the database search method of the above procedure, first, by using the database device having the data storage means, the storage means, and the search means, first, a read pattern for a plurality of data in the data storage means is set, This is stored in the data storage means. Next, a desired read pattern is designated from the read patterns stored in the data storage means. When the reading pattern is specified, only the data specified by the reading pattern is read from the data accumulating means and stored in the storing means, and the searching means extracts the data satisfying the search condition from the data in the storing means. To do. That is, according to the database search method, the data is extracted by the search means from only the data specified by the specified read pattern, not all the data in the data storage means.

[0018]

BEST MODE FOR CARRYING OUT THE INVENTION A database apparatus according to the present invention and a database search method in this database apparatus will be described below with reference to the drawings.

The database device of this embodiment is shown in FIG.
As shown in FIG. 1, the input unit 1, the output unit 2, and the secondary storage unit 3
And a search processing unit 4. However, it is assumed that this database device is an object-oriented database device in which the respective data making up the database are associated with each other by the object ID (see FIG. 5).

The input unit 1 is composed of, for example, a keyboard and a mouse, and is operated by a user to input object data for constructing a database, an instruction of a search condition when performing a data search, and a reading pattern described later. It is for specifying or setting. The output unit 2 is composed of, for example, a CRT and a printer, and is for displaying (outputting) the result of the data search to the user.

The secondary storage unit 3 is, for example, a magnetic disk or M
It is composed of O and accumulates a plurality of object data which constructs a database. That is,
The secondary storage unit 3 functions as data storage means in the present invention. The secondary storage unit 3 also has a function as a pattern storage unit in the present invention, and stores a plurality of read patterns set by the user from the input unit 1 and the like together with a plurality of object data. Has become.

The read pattern is composed of a set of object data set by a user or the like, and is composed of a set of object data in a specific field among the object data accumulated in the table format, for example. is there. This read pattern is
It is represented by a flag or the like corresponding to a specific field, and different read patterns are identified by the difference in this flag. Specifically, in the secondary storage unit 3, as shown in FIG. 2, the table-type data structure is considered three-dimensionally instead of two-dimensionally,
Different reading patterns in the three-dimensional direction,
The flag is stored so that it can be identified by the flag corresponding to the specific field.

In the secondary storage unit 3, it is not always necessary to store the object data in a three-dimensional table format. For example, a matrix showing the correspondence between the object data and the read pattern is stored separately from the object data. As a result, it may function as a pattern storage means. Further, the read pattern may be, for example, a set of object data in a specific record or a set of arbitrary object data in a table format, instead of a set of object data in a specific field.

In FIG. 1, the search processing unit 4 controls the entire database device and performs a data search for the object data stored in the secondary storage unit 3. Therefore, in the search processing unit 4, the control unit 4a
And a memory unit 4b and a calculation unit 4c.

The control unit 4a is composed of, for example, a CPU, interprets the command word received from the input unit 1 and the like, and also inputs the input unit 1, the output unit 2, the secondary storage unit 3, the memory unit 4b, and the arithmetic unit. By giving a control signal to the section 4c, the operation of these is controlled. Further, the control unit 4a recognizes the read pattern when the input unit 1 specifies the read pattern when performing the data search. Furthermore, the control unit 4a reads only the object data specified by the read pattern from the secondary storage unit 3 in accordance with the read pattern that has been recognized when performing the data search. That is, the control unit 4a has a function as a recognition unit and a reading unit in the present invention.

The memory unit 4b is composed of a semiconductor memory such as a RAM, and the control unit 4a has a secondary storage unit 3.
The object data read from is stored. That is, the memory section 4b functions as a storage means in the present invention. The calculation unit 4c is, for example, C
It is composed of a PU and extracts the object data that matches the search condition specified by the input unit 1 from the object data stored in the memory unit 4b by executing four arithmetic operations or logical operations. That is, the calculation unit 4c functions as the search means in the present invention.

Next, a database search method performed using the database device configured as described above will be described. However, here, the object data relating to “employee” in the table format as shown in FIG. 3 is accumulated in the secondary storage unit 3, and the object data in this table format has a data amount of one record of 38 bytes. And the data amount of all data is 532 bytes. Further, here, the memory unit 4b of the search processing unit 4
It is assumed that the data area has a capacity of 300 bytes.

In this database device, a read pattern is set in advance before data retrieval. That is, the user operates the input unit 1 to set a set of arbitrary object data among the object data accumulated in the secondary storage unit 3 as a read pattern. Then, the secondary storage unit 3 stores the set read pattern together with the object data. For example,
In the object data in the secondary storage unit 3, "Employee No.
, “Employee name”, “hire date”, “address”, etc., when a set of object data corresponding to fields such as “address” is set as one read pattern, the secondary storage unit 3 stores these objects. A set of data is made identifiable by a flag corresponding to each field (item), and its read pattern is registered as, for example, “A pattern”.

Here, when the user operates the input unit 1 to instruct a search condition and specify a read pattern, the search processing unit 4 in the database device follows the instruction and the specification, as shown in the flowchart of FIG. Search for data. For example, the user may say, “The date of joining the company is 4
If the search conditions such as “Employee No., employee name and address after the 1st of the month” are specified and “A pattern” is specified as the read pattern corresponding to these search conditions, the search processing unit 4
The control section 4a recognizes the specified read pattern "A pattern" (S101), and the data specified by the read pattern, that is, "employee No.
, “Employee name”, “hire date”, and “address” are read from the secondary storage unit 3 and stored in the memory unit 4b (S10).
2).

At this time, the data amount of these object data is 2 bytes for "employee No." and 4 for "employee name".
Bytes, "hire date" is 8 bytes, and "address" is 6 bytes. Therefore, the control unit 4a controls the object data ((2 bytes + 4 bytes + 8
(Byte + 6 bytes) × 14 = 280 bytes) can be read from the secondary storage unit 3 and stored in the memory unit 4b.

When the object data for all the records is stored in the memory unit 4b, the arithmetic unit 4c subsequently changes the object data stored in the memory unit 4b to "the date of joining the company is April 1, 1990. It is determined whether or not there is object data that matches the search conditions by comparing the search conditions such as “Employee No. after day and employee name and address” (S1).
03). As a result of this determination, if there is object data that matches the search condition, the search processing unit causes the object data to remain in the memory and swaps out object data that does not match the search condition. Then, when it is determined that all the object data in the memory unit 4b remain or are swapped out,
The control unit 4a sends the object data remaining in the memory as a data search result, and sends it to the output unit 2 for display to the user (S104).

As described above, in the database device of this embodiment and the data search method using this database device, only the object data specified by the read pattern is searched for in the secondary storage unit 3 when performing the data search.
From the memory unit 4b. Therefore, even if a large amount of object data is stored in the secondary storage unit 3, if the reading pattern is stored in advance and the reading pattern is designated at the time of data retrieval, the secondary storage unit 3 The data identified by the read pattern, not all the data in
That is, only necessary data will be read. As a result, it is possible to suppress the shortage of memory capacity in the memory unit 4b and frequent occurrence of swap-out, and it is possible to realize the saving of the memory capacity of the memory unit 4b, and as a result, improve the response of data search. be able to.

[0033]

As described above, the database device of the present invention reads only the data specified by the read pattern from the data storage means to the storage means when performing a data search. Therefore, even if a large amount of data is accumulated in the data accumulating means, if the reading pattern is stored in advance and the reading pattern is specified at the time of data retrieval, all the data in the data accumulating means will be stored. Instead, only the data specified by the read pattern will be read. As a result, in this database device, it is possible to suppress insufficient memory capacity and frequent occurrence of swap-out, and as a result, it is possible to improve the response of data search.

Further, according to the database search method of the present invention, when performing data search, only the data specified by the read pattern is read from the data storage means to the storage means. Therefore, even when a large amount of data is accumulated in the data accumulating means, the reading pattern is stored in advance and the reading pattern is specified at the time of data retrieval. Instead, only the data specified by the read pattern will be read. Therefore, if the data is searched by this database searching method, the same effect as described above can be obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an example of an embodiment of a database device according to the present invention.

FIG. 2 is an explanatory diagram showing a data structure accumulated in the database device of FIG.

FIG. 3 is an explanatory diagram showing a specific example of data accumulated in the database device of FIG.

FIG. 4 is a flowchart showing the procedure of a data search method performed using the database device of FIG.

FIG. 5 is an explanatory diagram showing data association by object ID in the object-oriented database device.

FIG. 6 is an explanatory diagram showing an example of a data structure accumulated in a conventional database device.

FIG. 7 is a flowchart showing a procedure of data search in a conventional database device.

FIG. 8 is an explanatory diagram illustrating swap-out.

[Explanation of symbols]

3 Secondary Storage Unit (Data Accumulation Means, Pattern Storage Means) 4a Control Unit (Identification Means, Reading Means) 4b Memory Unit (Storage Means) 4c Computing Unit (Search Means)

Claims (2)

[Claims] 1. A data storage means for storing a plurality of data, a storage means for storing the data read from the data storage means, and a data which meets a search condition from the data stored in the storage means. In a database device including a search unit, a pattern storage unit that stores a preset data set among a plurality of data stored in the data storage unit as a read pattern, and data extraction by the search unit Prior to the above, when a desired reading pattern is designated from the reading patterns stored in the pattern storing means, the recognizing means for recognizing the designated reading pattern and the recognizing means for recognizing the reading pattern are recognized. Only the data specified by the read pattern is read from the data storage means and stored in the storage means Database device characterized by comprising a reading means for. 2. A data storage means for storing a plurality of data, a storage means for storing the data read from the data storage means, and a data matching the search condition from the data stored in the storage means. A database search method using a database device comprising a search means, comprising: setting a set of arbitrary data among a plurality of data stored in the data storage means as a read pattern; A process of storing a pattern in the data storage unit together with the plurality of data, a process of designating a desired read pattern from the read patterns stored in the data storage unit, and only data specified by the designated read pattern. Is read from the data storage means and stored in the storage means A database search method comprising: a process; and a process of determining whether or not the data stored in the storage means meets the search condition.