JPH08166897A - Data storing method and device in database system - Google Patents

Abstract

PURPOSE: To efficiently perform the data storage processing in a data base system in short time. CONSTITUTION: An area securing part 2 secures the continuous unused page within a database file as an initial area and an overflow area and registers the attribute information specific to each area in an area information storage area 3. An initial address generation part 4 calculates the storage candidate page or the storage candidate address within the initial area by the key value of a processing object record and the attribute information of the initial area within the area information storage area 3. An overflow address generation part 5 calculates the storage candidate page or the storage candidate address within each overflow area by the key value of the processing object record and the attribute information of each overflow area within the area information storage area 3. A record processing execution part 6 performs a record storage processing by the storage candidate page or the storage candidate address calculated in the initial address generation part 4 and the overflow address generation part 5.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data storage method and data storage device in a database system,
In particular, by referring to the key value and the attribute information unique to the initial area or the overflow area, a database system that determines the storage destination page of the record storage candidate page in the initial area or the overflow area or the processing target record that is the storage candidate address The present invention relates to a data storage method and a data storage device.

[0002]

2. Description of the Related Art JP-A-61-216016 discloses
A technique related to a database system in which a region for storing a database is divided into a primary region composed of a plurality of pages and an overflow region for management is disclosed. As a method of managing pages in such an overflow area, a method of using an inter-page chain,
There is a method of preparing an overflow area in advance and managing which page is used in the overflow area.

In the former overflow page management method using a page chain, the top overflow page is linked from the storage candidate page by a page chain, and when there are a plurality of overflow pages, each overflow page is sequentially linked between pages. Link in a chain. In this method, in the record storing process, when the record to be processed cannot be stored in the storage candidate page obtained by hashing the key value, the chain between pages is sequentially traced, and there is a free space for storing the record in the overflow page. It is determined whether or not the record is stored.

On the other hand, in the latter method of preparing an overflow area in advance and managing an overflow page, when an overflow page is required, a page having an empty space in which a record to be processed can be stored is searched from within the area. , The found page is used as an overflow page.

[0005]

However, according to this conventional overflow page management method by the inter-page chain, when a record is added, another record is already stored in the storage candidate page to which the record to be processed is to be added. If there is no free space to store the record to be processed in the storage candidate page, or if the target record is not in the storage candidate page when searching for records, Has a problem that a page read process occurs for each page regardless of whether there is a target record or a free space because it is necessary to sequentially follow the chain between pages.

Further, since the inter-page chain is formed with the storage candidate page as a starting point, even if an attempt is made to simultaneously process a plurality of records having different key values so that the hashing result is the same storage candidate page, the overflow page is processed. Therefore, there is a problem regarding the simultaneous running property that the page exclusive control process occurs and other processes cannot operate until one record process is completed.

In the conventional overflow page management method using the page usage status in the overflow area, the overflow page is managed regardless of the key value of the record to be processed. Since it is necessary to search for all the overflow pages used in the area while judging whether or not they are the target records, there is a problem that the page reading process occurs more than necessary and the processing time at the time of search increases. .

An object of the present invention is to efficiently execute data storage processing in a database system in a short time.

Another object of the present invention is to sequentially store the storage addresses of the records in the overflow page by tracing the inter-page chain even when a synonym occurs when a record is added to the database and the page in the initial area overflows. Rather than searching for it, it was necessary to trace the inter-page chain by calculating and deciding the storage address in each overflow area based on the key value and overflow area information, as in the initial area. This is to reduce the number of page readings and shorten the processing time.

Still another object of the present invention is to calculate a storage address using attribute information unique to a secured area, so that a record group that has become a synonym in the initial area or a synonym in an overflow area can be identified. This is to improve the simultaneous running property of the record processing by storing the record groups that have been stored in the next prepared overflow area by distributing them to different pages.

[0011]

According to a first aspect of the present invention, there is provided a data storage method in a database system, which comprises a search step of searching for continuous unused pages in a database file, and a securing step of securing an initial area and an overflow area. When the record is stored in the initial area and the record cannot be stored in the initial area,
A storage step of storing in the overflow area.

The data storage device in the second database system of the present invention uses a designated field in a record as a key item, and determines a storage address for the database file of the record based on the key value which is the value of the field. A data storage device in the system, which secures an area for searching consecutive unused pages in the database file and securing them as an initial area and an overflow area, and registering attribute information unique to the secured area in the area information storage area. And a storage candidate page or storage candidate address in the initial area is calculated using the key value of the record to store the record in the initial area secured by the area securing unit. , The record to the area pointed to by either the page or the address When the record cannot be stored in the initial area while storing the record, the overflow area is stored using the key value of the record in order to store the record in the overflow area secured by the area securing unit. And a control unit for calculating one of the storage candidate page and the storage candidate address and storing the record in an area designated by one of the page and the address.

The data storage device in the third database system of the present invention uses a designated field in a record as a key item, and determines a storage address for the database file of the record based on the key value which is the value of the field. A data storage device in the system, which secures an area for searching consecutive unused pages in the database file and securing them as an initial area and an overflow area, and registering attribute information unique to the secured area in the area information storage area. A storage candidate in the initial area based on a copy number, the key value of the record, and the attribute information of the initial area secured by the area securing unit registered in the area information storage area by the region securing unit. An initial address generation unit that calculates either a page or a storage candidate address, Either a storage candidate page or a storage candidate address in the overflow area based on the key value of the record and the attribute information of the overflow area secured by the area securing unit registered in the area information storage area. Record processing is executed to store the record in an area indicated by either an overflow address generation unit that calculates one, the initial address generation unit, or the storage candidate page or the storage candidate address calculated by the overflow address generation unit. Section, in order to store the record in the initial area, the initial address generation section is activated to calculate either a storage candidate page or a storage candidate address in the initial area, and the record processing execution section Start up and store the record in the area pointed to by either the page or the address If the record cannot be stored in the initial area, either the storage candidate page or the storage candidate address in the overflow area is activated by activating the overflow address generation unit to store the record in the overflow area. And a control unit for calculating one of them, activating the record processing execution unit, and storing the record in an area designated by either the page or the address.

The data storage device in the fourth database system of the present invention uses a designated field in a record as a key item, and determines a storage address for the database file of the record based on the key value which is the value of the field. A data storage device in the system, which is an area reserve unit that searches for consecutive unused pages in the database file and secures it as an initial region and an overflow region, and an initial region and an overflow region that are secured by this region reserve unit. An area information storage area for registering group data consisting of an area number, the number of pages in the area, and a first page number as attribute information, a key value of the record, and the area reservation unit registered in the area information storage area. Based on the attribute information of the initial area secured by the area securing unit It is ensured by an initial address generation unit that calculates either a storage candidate page or a storage candidate address in the initial area, and a key value of the record and the area reservation unit registered in the area information storage area. An overflow address generation unit that calculates either a storage candidate page or a storage candidate address in the overflow region based on the attribute information of the overflow region, the initial address generation unit, and the overflow address generation unit. In the case where the attribute information of the initial area is not stored in the area indicated by either the calculated storage candidate page or the storage candidate address, the record processing execution unit that stores the record, and the area information storage area ,
The area allocation unit is activated to secure the initial area, the attribute information of the initial area is registered in the area information storage area, and the initial address generation unit is activated to store the record in the initial area. One of the storage candidate page and the storage candidate address in the initial area is calculated, and the record processing execution unit is activated to store the record in the area pointed to by either the page or the address. If the record could not be stored in the initial area and the attribute information of the overflow area is not stored in the area information storage area, the area securing unit is activated to secure the overflow area. However, in order to register the attribute information of the overflow area in the area information storage area and store the record in the overflow area, the overflow address is stored. The generation unit is activated to calculate either the storage candidate page or the storage candidate address in the overflow area, and the record processing execution unit is activated to move the record to the area indicated by either the page or the address. When the record cannot be stored in the overflow area and the attribute information of another overflow area is not stored in the area information storage area, the area securing unit is activated. Secures another overflow area, registers the attribute information of the other overflow area in the area information storage area, and activates the overflow address generation unit to store the record in the other overflow area. Calculate either storage candidate page or storage candidate address in the area,
And a control unit for activating the record processing execution unit and storing the record in an area designated by either the page or the address.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below in detail with reference to the drawings.

Referring to FIG. 1, a data storage device in a database system according to an embodiment of the present invention searches for consecutive unused pages in a database file,
An area securing unit 2 that secures this as an initial area or an overflow area, registers attribute information unique to the secured area in the area information storage area 3, and at the time of record storage,
An initial address generation unit 4 that calculates a storage candidate page or a storage candidate address in the initial area based on the key value of the processing target record and the attribute information of the initial area registered in the area information storage area 3, and the processing target. An overflow address generation unit 5 for calculating a storage candidate page or a storage candidate address in each overflow area based on the key value of the record and the attribute information of each overflow area registered in the area information storage area 3, and an initial address Generation unit 4,
And a record processing execution unit 6 that performs a record storage process based on the storage candidate page or the storage candidate address calculated by the overflow address generation unit 5.

Referring to FIG. 4, the area information recording unit 3 is
Initial area information 301 storing attribute information of the initial area,
It is composed of overflow area information 302 for storing attribute information of the overflow area and a page structure 303.

The initial area information 301 includes the area number, the number of pages in the area, and the first page number as attribute information. In the example of FIG. 4, the continuous page having the initial area number i is the page number x in the database file. It means that s pages have been secured.

The overflow area information 302 includes the area number, the number of pages in the area, and the first page number as attribute information. In the example of FIG. 4, continuous pages having the overflow area number 1 are from page number y to page t. This shows that the continuous pages having the overflow area number 2 are secured for the page numbers z to u.

The page structure 303 is the initial area information 301.
And the overflow area information 302 are associated with each other.

Next, a data storage device in a database system according to an embodiment of the present invention will be described with reference to FIGS.
Will be described with reference to.

In response to a record storage request from the user,
The control unit 1 first determines whether or not the attribute information about the initial area is registered in the area information recording unit 3 (step 10
1) If it is determined that it is registered, step 1
To 04.

When it is determined that the attribute information of the initial area is not registered in the area information recording section 3, the control section 1 activates the area securing section 2 (step 102).

The area reserving unit 2 retrieves an unused continuous page from the database file and reserves it as an initial area (step 201), and registers information on the secured initial area in the area information recording area 3 as attribute information. (Step 202).

Next, the controller 1 activates the initial address generator 4 (step 103).

The initial address generation unit 4 obtains the key value of the record to be processed from the control unit 1 and the attribute information of the initial area from the area information storage area 3, and calculates the storage candidate address in the initial area. , Control is returned to the control unit 1.

The processing of the initial address generator 4 will be described in detail below.

The initial address generator 4 acquires the key value of the record to be processed from the controller 1 (step 40).
1) The initial area information 301, which is the attribute information of the initial area, is acquired from the area information storage area 3 (step 402).

Next, the initial address generator 4 determines the area number in the initial area information 301, the top page number, and the controller 1
A computed value to be hashed is generated from the key value of the processing target record obtained from (step 403).

The method of generating the operand value is, for example, a region number,
There is a method of weighting in order of the first page number and the key value and taking the sum.

The initial address generator 4 operates in step 403.
The value obtained in step 2 is hashed (step 404), and the storage candidate page number in the initial area is calculated from the value of the hashing result and the number of pages in the area in the initial area information 301 (step 405).

As a method of calculating the storage candidate page number, for example, there is a method of adding the first page number of the initial area to the remainder when the value of the hash result is divided by the number of pages.

When the processing of the initial address generation unit 4 is completed, the control unit 1 activates the record processing execution unit 6 (step 104).

The record processing execution unit 6 reads from the database file the corresponding page in the initial area pointed to by the storage candidate address calculated by the initial address generation unit 4 (step 601), and whether the record to be processed can be stored in that page. It is determined whether or not (step 602), and when it is determined that the processing target record can be stored, the processing target record is stored in the page (step 603).

When the record processing execution unit 6 determines that the record to be processed cannot be stored in the page, it returns the process to the control unit 1.

When the control unit 1 is notified by the record processing execution unit 6 that the record cannot be stored at the storage candidate address in the initial area, the overflow area information 302, which is attribute information about the overflow area, is registered in the area information recording unit 3. It is determined whether or not it is registered (step 105), and when it is determined that it is registered, the process proceeds to step 108.

When it is determined in step 105 that the overflow area information 302 is not registered, the control section 1 activates the area securing section 2 (step 10).
6).

The area reserving unit 2 retrieves an unused continuous page from the database file and reserves it as an overflow area, as in the case of reserving the initial area (step 203).
At the same time, information about the secured overflow area is registered in the area information recording area 3 as attribute information, that is, overflow area information 302 (step 204), and the process is returned to the control unit 1.

Next, the controller 1 activates the overflow address generator 5 (step 107).

The overflow address generation unit 5 acquires the key value of the record to be processed from the control unit 1 and the overflow area information 302 from the area information storage area 3, calculates the storage candidate address in the overflow area, and controls it. To the control unit 1.

The processing of the overflow address generator 5 will be described in detail below.

The overflow address generation unit 5 acquires the key value of the record to be processed from the control unit 1 (step 501) and acquires the overflow area information 302, which is the attribute information of the overflow area, from the area information storage area 3. (Step 502).

Next, the overflow address generation unit 5 uses the area number in the overflow area information 302, the first page number, and the key value of the processing target record obtained from the control unit 1 to calculate the value to be hashed. Is generated (step 503).

As the method of generating the operand, a weighting method may be adopted as in the processing in the initial address generating section 4, or another method may be set for overflow address generation. In any case, the one that can obtain a value different from the value to be calculated at the time of generating the initial address is adopted.

The overflow address generator 5 hashes the value obtained in step 503 (step 50).
4), value of hashing result and overflow area information 30
The storage candidate page number in the overflow area is calculated from the number of pages in area 2 (step 505).

As a method of calculating the storage candidate page number, the same method as that at the time of generating the initial address can be considered.

When the processing of the overflow address generation section 5 is completed, the control section 1 restarts the record processing execution section 6 (step 108).

The restarted record processing execution unit 6 reads the corresponding page in the overflow area indicated by the storage candidate address calculated by the overflow address generation unit 5 from the database file (step 604), and the record to be processed is the page. It is determined whether or not the record can be stored in the page (step 605), and when it is determined that the record can be stored in the page, the record to be processed is stored in the page (step 60).
6).

If the record processing execution unit 6 determines that the record to be processed cannot be stored in the page, it cannot store the record to be processed in the overflow area, so that the record to be processed can be stored in the initial area. In the same manner as in the case where there is no such a case, the processing is returned to the control unit 1, the overflow area is secured again by the area securing unit 2, and the storage candidate address for another newly secured overflow area is calculated by the overflow address generating unit 5 and then re-repeated. Each time, the record processing execution unit 6 is activated to store the record.

By this series of processing, the records having different key values stored in the same page in the initial area are
In the overflow area, it will be stored in another page.

For example, in FIG. 4, the key value, the area number i of the initial area information 301, the number of pages in the area s, and the start page number x are stored in the overflow area 1 in order to find the storage candidate page in the initial area. In order to obtain the candidate address, the key value, the area number j of the overflow area information 302, the number of pages in the area t, and the top page number y are set as variables.
Further, by using the key value, the area number k of the overflow area 302, the number of pages in the area u, and the first page number z as variables for obtaining the storage candidate address in the overflow area 2, the page number "j" is set in the initial area. The records having the key values “a”, “b”, “c”, and “d” stored in “” are stored in the page number “k” in the overflow area 1 as the key values “a” and “b”. Only records having the key value "a" can be stored in the page number "l" in the overflow page 2, and a record group in which a synonym occurs in the initial area is stored. Can be distributed in the overflow region.

For example, as shown in FIG. 4, in the initial area information 301 in the area information storage area 3, the initial area number i, the number of pages in the area s, and the initial area indicated by the first page number x are registered, and the overflow area is registered. An overflow area indicated by an overflow area number j, the number of pages in the area t, and a start page number y is registered in the information 302, and further, an overflow area number k, the number of pages in the area u, and a start page number z are registered in the overflow area 302. It is assumed that the overflow area is registered.

First, the record having the key value a is set to the initial area i.
Since the page j is shown when the storage candidate address is calculated by the initial address generation unit 4 to store it in the page, the record processing execution unit 6 stores the record in the page j. Similarly, key value b, key value c, and key value d
When the storage candidate address in the initial area i is calculated for each of the records with, the page j is shown as in the case of the key value a, and the record processing execution unit 6 stores the record.

Next, the record having the key value a is set to the initial area i.
The page j is shown when the storage candidate address is calculated by the initial address generation unit 4 to store it in the first area. However, since the record cannot be stored in the page j of the initial area i, the overflow is attempted to be stored in the overflow area 1. When the storage candidate address is calculated by the address generation unit 5, page k is shown. Therefore, the record processing execution unit 6 stores the record in page k. Similarly, an attempt to store a record having the key value b in page j of the initial area i indicated by the initial address generator 4 cannot be performed, so that the overflow address generator 5 attempts to store the record in the overflow area 1 and the storage candidate is stored. When the address is calculated, page k is shown, and the record is executed by the record processing execution unit 6. In the same way,
The overflow address generation unit 5 attempts to store a record having the key value c and the key value d in the overflow area 1.
When the storage candidate address is calculated by, the page m is shown instead of the page k, and the record processing execution unit 6 stores the record.

Further, in order to store the record having the key value a in the initial area i, page j is shown when the storage candidate address is calculated by the initial address generating unit 4, but the page j of the initial area i has the record. Could not be stored, and page overflow k was shown when the storage address candidate was calculated by the overflow address generation unit 5 in order to store in the overflow area 1. However, since this record cannot be stored in page k either, it is stored in the overflow area 2. When the storage address is calculated by the overflow address generation unit 5 in order to indicate the page 1, the record processing execution unit 6 stores the record in the page 1. Similarly, when a storage candidate address is calculated by the overflow address generation unit 5 in an attempt to store the record having the key value b, the key value c, and the key value d in the overflow area 2, respectively, page o, page n, and page Since p is indicated, the record processing execution unit 6 stores the record.

As described above, by changing the calculation method of the storage candidate address for each key value or for each attribute information of each area, records having different key values stored in the same page in the initial area overflow. In the area, it will be stored in another page.

With the above, the processing of the data storage device in the database system according to the embodiment of the present invention is completed.

In the data storage device in the database system according to the embodiment of the present invention, even if a page in the initial area overflows due to the occurrence of a synonym when a record is added to the database, sequential overflow pages are traced in the inter-page chain. Instead of searching for the storage address of the record inside, the inter-page chain is calculated by calculating and deciding the storage address in each overflow area based on the key value and overflow area information as in the initial area. It has the effect of reducing the number of page reads required for tracing and shortening the processing time. Furthermore, by calculating the storage address using the attribute information unique to the secured area, Records that were synonyms in the initial area, or synonyms in the overflow area The stomach record group, has been in the overflow area available then each store is dispersed in another page, and has an advantage of being able to improve the simultaneous running of record processing.

[0059]

As described above, the data storage device in the database system of the present invention has an effect that the data storage processing in the database system can be efficiently executed in a short time.

Further, even when a synonym occurs when a record is added to the database and the page in the initial area overflows, the storage address of the record is not sequentially searched for in the overflow page by tracing the chain between pages. As with the initial area, by calculating and deciding the storage address in each overflow area based on the key value and overflow area information, the number of page read times required to follow the inter-page chain is reduced. It has an effect that the processing time can be shortened.

Further, by calculating the storage address using the attribute information peculiar to the secured area, the record group that is a synonym in the initial area or the record group that is a synonym in the overflow area is Next, in the prepared overflow area, the pages are separately distributed and stored in different pages, which has the effect of improving the simultaneous running property of the record processing.

[Brief description of drawings]

FIG. 1 is a block diagram showing one embodiment of the present invention.

FIG. 2 is a flowchart showing the processing of the control unit 1, the area securing unit 2, and the initial address generation unit 4 in the embodiment of the present invention.

FIG. 3 is a flow chart showing processing of an overflow address generation unit 5 and a record processing execution unit 6 in an embodiment of the present invention.

FIG. 4 is a diagram showing the contents of an area information storage area 3 in an embodiment of the present invention.

[Explanation of symbols]

 1 control section 2 area reservation section 3 area information storage area 4 initial address generation section 5 overflow address generation section 6 record processing execution section

Claims (3)

[Claims] 1. A retrieval step of retrieving consecutive unused pages in a database file, a securing step of securing as an initial area and an overflow area, a record is stored in the initial area, and the record is stored in the initial area. If not possible, a storage step of storing in the overflow area is provided, and a data storage method in a database system. 2. A data storage device in a database system, wherein a designated field in a record is used as a key item, and a storage address for the database file of the record is determined based on a key value which is the value of the field. The area reservation unit that searches for consecutive unused pages in the area and reserves them as the initial area and the overflow area, and registers the attribute information unique to the secured area in the area information storage area, and the area reservation unit that secures the record. In order to store it in the initialized initial area, either the storage candidate page or the storage candidate address in the initial area is calculated using the key value of the record, and either the page or the address is pointed to. The record cannot be stored in the initial area while storing the record in the area In this case, in order to store the record in the overflow area secured by the area securing unit, either the storage candidate page or the storage candidate address in the overflow area is calculated using the key value of the record. And a control unit for storing the record in an area designated by either the page or the address. 3. A data storage device in a database system, wherein a designated field in a record is used as a key item, and a storage address for the database file of the record is determined based on a key value which is the value of the field. Area reservation unit that searches for consecutive unused pages in the area and reserves them as the initial area and overflow area, and registers the attribute information unique to the secured area in the area information storage area, the key value of the record and the area reservation Calculates either a storage candidate page or a storage candidate address in the initial area based on the attribute information of the initial area secured by the area securing section registered in the area information storage area An initial address generation unit, and a key value of the record and before being registered in the area information storage area An overflow address generation unit that calculates one of a storage candidate page or a storage candidate address in the overflow region based on attribute information of the overflow region secured by the storage region securing unit; and the initial address generation unit, And a record processing execution unit that stores the record in an area indicated by either the storage candidate page calculated by the overflow address generation unit or the storage candidate address, and the record processing execution unit that stores the record in the initial area. The initial address generation unit is activated to calculate either the storage candidate page or the storage candidate address in the initial area, and the record processing execution unit is activated to the area pointed to by either the page or the address. While storing the record, if the record cannot be stored in the initial area, In order to store the code in the overflow area, the overflow address generation unit is activated to calculate either a storage candidate page or a storage candidate address in the overflow area, and the record processing execution unit is activated to start the page. A data storage device in a database system, comprising: a control unit for storing the record in an area indicated by either one of the addresses.