JPH05334146A - Database management system - Google Patents

Abstract

PURPOSE:To reduce the load on a secondary storage device and to easily repair a table at the time of occurring a fault by reducing the number of files to be prepared in the secondary storage device in the case of distributing the table to plural secondary storage devices. CONSTITUTION:This database managing system 30 is provided with a table area distributing part 32 for managing the distributed state of a table in addition to a file area managing part 33 for managing the arrangement or the like of the table on respective secondary storage devices 34, 35. The distributing part 32 is provided with striping management information 40 for managing data for identifying a file storing data in the preceding processing in each table, and at the time of tuple inserting processing, an idle area is retrieved based on the information of the distributing part 32, and in the case of newly securing an idle area, file change is executed to distribute the table to plural files 36, 37.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database management system.

[0002]

2. Description of the Related Art In a database management system, data is managed in the form of a table which is a set of tuples (usually called records, which is data for one row). Further, even when the access to the table is concentrated, the load on the secondary storage device that stores the table is reduced and a high-speed response is realized. Therefore, normally, the table is distributed to a plurality of files,
Each file is stored in a separate secondary storage device. The process of dividing the table into a plurality of files and distributing / storing the files in the secondary storage device is called a striping process of the table.

FIG. 2 shows a schematic configuration of a conventional database management system. In FIG. 2, reference numeral 10 is a database management system, 11 is a file area management unit equipped in the database management system 10,
12 and 13 are secondary storage devices, and 14 is a secondary storage device 1.
2 is a database file stored in 2, and 15 is a database file stored in the secondary storage device 13.

Each of the files 14 and 15 described above is a distribution of one table. Further, the file area management unit 11 in the database management system 10 manages how the files 14 and 15 are stored in the secondary storage devices 12 and 13, and when a tuple is newly inserted in the table. First, the striping process is performed according to the state of the free area in the managed file.

Next, the tuple insertion processing in the database management system shown in FIG. 2 will be described with reference to FIG. Note that the tuple insertion instruction is issued by the secondary storage device 1.
It is assumed that the file is received while referring to the file 14 stored in 2.

When a tuple insertion command is received, it is first determined whether or not there is a free space in the acquired data storage area (file area) on the file 14 currently being referenced (step 21), and there is a free space. In this case, the tuple is stored in the empty area (step 25). However, if it is determined in step 21 that there is no free space, another data storage area in the file 14 being referred to is acquired (step 22). Then, it is judged whether or not there is a free space in the newly acquired data storage area (step 23), and if there is no free space, the file to be referred to is switched to the next file (file 15 in this example), and The process returns to step 22 (step 24). If it is determined in step 23 that there is a space, the process proceeds to step 25, where a tuple is inserted in the empty area and the process ends.

That is, the conventional database management system simply performs a striping process on the table by creating a file on the next secondary storage device when the data storage area is insufficient. The files will be distributed to all the secondary storage devices installed in the system. The distributed table is stored in the file area management unit 11
It is collectively managed by management of memory addresses, etc.

[0008]

However, in the case of the conventional database management system described above, when there are a plurality of tables constituting the database, the system is equipped with a plurality of files corresponding to the respective tables. The number of actual files to be managed is doubled in proportion to the number of secondary storage devices because each secondary storage device must be created. Therefore, since the number of file information that each process can have is limited by the operating system, the number of striping processes that can be performed is limited.

Further, by striping processing, each 2
As many files as the number of tables that make up the database are created on the secondary storage device, and when a failure occurs in the secondary storage device, the number of tables subject to the failure becomes huge. There was also a problem.

The present invention has been made in view of the above circumstances, and it is possible to reduce the number of files to be created on the secondary storage device by the striping process, and in the case where a failure occurs in the secondary storage device. Also reduces the number of tables subject to failure, and reduces the load on the secondary storage device when access to the table is concentrated, and at the same time, it is easy to repair the table when the secondary storage device fails. The purpose is to provide a secure database management system.

[0011]

A database management system according to claim 1 is a database management system for managing tabular data by distributing it to a plurality of secondary storage devices, each of which has two memory addresses or the like. In addition to the file area management unit that manages the layout of the table on the next storage device, a table area distribution unit that manages the distribution status of the table is provided.

The table area distribution unit is equipped with striping management information for managing the data for identifying the file in which the data was previously stored for each table, and the table area distribution is performed during the tuple insertion process. The table is distributed to a plurality of files by searching for an empty area based on the information of the copy and executing a file change when a new empty area is secured.

A database management system according to a second aspect is a further improvement of the database management system according to the first aspect, wherein the striping management information indicates whether a table to be processed is to be distributed. The striping target table data shown is provided, and the distribution is executed only in the table set as the distribution target in the striping target table data.

[0014]

According to the database management system of the present invention, in addition to the file area management section for managing the table on each secondary storage device by the memory address or the like, the table area distribution section for managing the distribution status of the table. It is equipped with. The file in which the table data is stored last time can be identified by the striping management information of the table area distribution unit, and at the time of tuple insertion, the secondary storage device having a free space based on the striping management information of the table area distribution unit. Striping is performed by selecting and creating a file.

Therefore, the file for distributing the table is
It is possible to select from all the secondary storage devices installed in the system, and it is not necessary to create each of the secondary storage devices installed in the system. In other words, when there are a plurality of tables forming the database, it is not necessary to create a plurality of files corresponding to each of the tables for each secondary storage device installed in the system. It is possible to distribute a table to a plurality of secondary storage devices only by equipping one secondary storage device with at least one database file, and it should be managed in proportion to the number of secondary storage devices. It is possible to prevent the inconvenience of doubling the number of files in the secondary storage device and reduce the number of files created in the secondary storage device. Therefore, the number of file information that each process can have is limited by the operating system. However, you can increase the number of striping processes within that limit,
It is possible to reduce the load on the secondary storage device when access to the table is concentrated.

Further, since the number of files created on one secondary storage device itself is reduced, the number of tables subject to the failure is reduced even when a failure occurs in the secondary storage device. Repair is also easy.

Further, in the database management system according to the second aspect, it is possible to further limit the tables for which the striping process is executed by the striping management table data of the striping management information, and only perform the striping process in advance for the tables where access is concentrated. By specifying it as the target of, it is possible to avoid unnecessary striping processing and operate the database more efficiently.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows a schematic configuration of an embodiment of a database management system according to the present invention. In FIG. 1, reference numeral 30 is a database management system, 3
Reference numeral 1 is a table area management unit equipped in the database management system 30, 34 and 35 are secondary storage devices, and 36, respectively.
Is a database file stored in the secondary storage device 34, and 37 is a database file stored in the secondary storage device 35.

Each of the files 36 and 37 described above is a distribution of one table. Further, the table area management unit 31 in the database management system 30 is configured to include a table area distribution unit 32 and a file area management unit 33 as illustrated. Here, the table area distribution unit 32 has striping management information 40 as information for managing the status of striping processing for various tables that are constituent elements of the database managed by the database management system 30. This striping management information 40
As shown in FIG. 4, the striping target table data 41 enumerating the tables to be subjected to striping processing among the various tables that are the constituent elements of the database managed by the database management system 30 and the striping Each table listed in the target table data 41 is composed of previously stored file identifier data 42 in which the identifier of the file stored in the previous data processing is described.
In addition, in FIG. 4, a table regarding employees and a table regarding business partners are illustrated as the tables to be subjected to the striping process. It is assumed that the data is stored in the file with the identifier “0”, and the table regarding the business partner is stored in the file with the identifier “2” at the time of the previous data storage. Also, the table area distribution unit 3
In 2, even for non-striping table,
It manages the file identifier that stored the data last time. Further, the table area distribution unit 32 is provided with a file identifier holding function 43 for holding a file identifier in a predetermined storage unit at the time of tuple insertion processing or the like.

The file area management unit 33 determines, as in the conventional case, how the files 36 and 37 are arranged in the respective secondary storage devices 34 and 35, which are system constituent elements, at the memory address. Etc.

In the database management system of this embodiment, when a tuple is newly inserted in the table, based on the striping management information 40 of the table area distribution unit 32,
Striping processing is performed according to the free space status in the managed file.

Next, tuple insertion processing in the database management system in one embodiment will be described with reference to FIG. It is assumed that the tuple insertion instruction is received while referring to the file 36 stored in the secondary storage device 34.

When a tuple insertion command is received, it is first determined whether or not there is a free space in the acquired data storage area (file area) of the file 36 currently being referenced (step 51), and there is a free space. In this case, the tuple is stored in the empty area and the process is terminated (step 59). However, if it is determined in step 51 that there is no free space, the file identifier holding function 43 clears the file identifier held in the predetermined storage means (step 52), and the striping of the table area distribution unit 32 is performed. A file identifier for storing the table is obtained based on the management information 40 (step 53).

Next, the file identifier obtained in step 53 is the same as the file identifier held in the predetermined storage means by the process of step 55 (that is, the file identifier storing process by the file identifier holding function 43) in the previous process. It is determined whether or not (step 5)
4). If they are not the same, the file identifier obtained in step 53 is used as the file identifier holding function 4 described above.
3 is stored in a predetermined storage means (step 5).
5).

Next, it is determined whether or not the table into which the tuple is to be inserted is the striping management information based on the striping management information 40. If the table is the striping processing target, the previously stored file identifier data of the striping management information 40 is determined. The file identifier held in 42 is updated to the next file identifier (step 57), and then
Go to step 58. If it is determined in step 56 that the table is not the striping process target table, the process immediately proceeds to step 58.

In step 58, it is judged whether or not the file having the file identifier obtained in step 53 has a free area, and if there is a free area, the process proceeds to the next step 59. If there is no free area, the process proceeds to step 53. Return and repeat the process.

If it is determined in step 54 that the file identifier is the same as the previous file identifier, the tuple insertion command is notified that there is no free space, and the process ends (step 60).

That is, in the database management system of one embodiment, in addition to the file area management unit 33 that manages the table on each secondary storage device by the memory address and the like, the table area distribution for managing the distribution status of the table is also provided. The unit 32 is provided. Then, the file in which the table data is stored last time can be identified by the striping management information 40 of the table area distribution unit 32, and at the time of tuple insertion, there is a space based on the striping management information 40 of the table area distribution unit 32. Striping processing is performed by selecting a secondary storage device and creating a file.

Therefore, the file for distributing the table is
It is possible to select from all the secondary storage devices installed in the system, and it is not necessary to create each of the secondary storage devices installed in the system.

In other words, when there are a plurality of tables constituting the database, it is not necessary to create a plurality of files corresponding to the respective tables for each secondary storage device equipped in the system. For example, it is possible to distribute a table to a plurality of secondary storage devices by equipping one secondary storage device with at least one file for a database, and to manage in proportion to the number of secondary storage devices. It is possible to prevent the inconvenience of doubling the number of existing files that should be present and reduce the number of files created in the secondary storage device. Therefore, the number of file information that can be held for each process depends on the operating system. Even if there is a limit, you can increase the number of striping processes that can be performed within that limit. It is possible to reduce the load.

Further, by reducing the number of files themselves created in one secondary storage device, the number of tables subject to the failure is reduced even when a failure occurs in the secondary storage device. Repair is also easy.

Further, in the above-described embodiment, the striping target table data 41 of the striping management information 40 can limit the tables on which the striping process is executed, and only the tables in which access is concentrated are designated in advance as the targets of the striping process. By doing so, it becomes possible to avoid unnecessary striping processing and operate the database more efficiently.

[0033]

According to the database management system of the first aspect, in addition to the file area management unit for managing the table on each secondary storage device by the memory address and the like, a table area for managing the distribution status of the table. It has a distributor. The file in which the table data is stored last time can be identified by the striping management information of the table area distribution unit, and at the time of tuple insertion, the secondary storage device having a free space based on the striping management information of the table area distribution unit. Striping is performed by selecting and creating a file.

Therefore, the file for distributing the table is
It is possible to select from all the secondary storage devices installed in the system, and it is not necessary to create each of the secondary storage devices installed in the system. In other words, when there are a plurality of tables forming the database, it is not necessary to create a plurality of files corresponding to each of the tables for each secondary storage device installed in the system. It is possible to distribute a table to a plurality of secondary storage devices only by equipping one secondary storage device with at least one database file, and it is necessary to manage in proportion to the number of secondary storage devices. It is possible to prevent the inconvenience of doubling the number of files in the secondary storage device and reduce the number of files created in the secondary storage device. Therefore, the number of file information that each process can have is limited by the operating system. However, you can increase the number of striping processes within that limit,
It is possible to reduce the load on the secondary storage device when access to the table is concentrated.

Further, since the number of files created on one secondary storage device itself is reduced, the number of tables subject to the failure is reduced even when a failure occurs in the secondary storage device. Repair is also easy.

Further, in the database management system according to the second aspect, it is possible to further limit the table on which the striping process is executed by the striping management table data of the striping management information, and only the table in which the access is concentrated is previously striped. By specifying it as the target of, it is possible to avoid unnecessary striping processing and operate the database more efficiently.

[Brief description of drawings]

FIG. 1 is a schematic configuration of an embodiment of the present invention.

FIG. 2 is a schematic configuration of a conventional database management system.

FIG. 3 is a flowchart showing a process when inserting a tuple in a conventional system.

FIG. 4 is an explanatory diagram of striping management information according to an embodiment of the present invention.

FIG. 5 is a flowchart showing a process when inserting a tuple according to the embodiment of the present invention.

[Explanation of symbols]

 30 database management system 31 table area management unit 32 table area distribution unit 33 file area management unit 34, 35 secondary storage device 36, 37 file 40 striping management information 43 file identifier holding function

Claims (2)

[Claims] 1. A database management system for managing tabular data by distributing it to a plurality of secondary storage devices, and managing file areas for managing the layout of tables on each secondary storage device by memory addresses and the like. In addition to the section, the table area distribution unit for managing the distribution status of the table is provided, and the table area distribution unit manages the data for identifying the file storing the previous data for each table by striping. Management information must be provided, and when tuple insertion processing is performed, a free area is searched based on the information in the table area distribution unit, and a file change is executed when a new free area is secured. A database management system that distributes tables to multiple files in. 2. The striping management information is equipped with striping target table data indicating whether or not a table to be processed becomes a target of distribution, and a table set as a target of distribution in the striping target table data is provided. The database management system according to claim 1, wherein the distribution is executed only for a limited time.