JP2658969B2 - Data storage control system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control system for initially storing data records in a storage destination database file from a storage source database file, and more particularly to a data storage control system for setting a free area for each cluster.

[0002]

2. Description of the Related Art Conventionally, this type of data storage control system has been used to set a free space in a database in consideration of the fact that new data is added after initial storage of data in a storage destination database file. ing.

[0003] As is well known, a database is a collection of data necessary and sufficient to fulfill a certain purpose. The database is generally stored as a database file in an auxiliary storage device, but a direct access storage device (DASD) is often used as the auxiliary storage device. A direct access storage device is an auxiliary storage device whose access time does not vary depending on the storage location of data. A virtual storage access method (VSAM) is known as a method for accessing a direct access storage device. The virtual storage access method is an access method capable of directly or sequentially processing fixed-length records and variable-length records on a direct access storage device.

In a database file, a data set is stored as a set of records. Generally, a record having a logical relationship with the root record is located near a root record, which is the first record, in a storage area. Is done. A collection of a root record and records having a logical relationship with the root record is called a cluster.

One example of a conventional data storage control system is:
It is described in JP-A-5-250234. FIG.
FIG. 5 shows a data record storage control method described in this publication, and FIG. 5 shows a publicly known technique. 4 and 5, a double circle “◎” indicates a root record, a single circle “○” indicates a record, a broken line “…” indicates a logical connection, and a black circle “●” indicates an additional record. . 4 and 5, (a) shows a state at the time of initial data storage in which a free area is set, and (b) shows a state in which new data (record) is added after the initial data storage.

According to the data record storage control method described in the above publication, as shown in FIG. 4A, the size of a free area is calculated based on the data amount of a cluster C, and after storing the cluster, the free space is calculated in block units. Set the area. In this type of data storage control method, as shown in FIG.
Suitable for adding a large number of logically connected records.

[0007] The publicly known technique can secure a certain free space in a block for all blocks in a database file at the time of data storage, as shown in FIG. is there. As shown in FIG. 5B, this type of data storage control method is suitable for a case where there is no logical connection between the additional records ● and the records are added on average in the cluster Ca.

[0008]

The prior art shown in FIGS. 4 and 5 has the following problems. That is,
When initially storing a storage source database file in which clusters having different characteristics of added data coexist in a storage destination database file, a conventional data storage control method for setting an empty area in block units after cluster storage (FIG. 4) ) Or a data storage control method (FIG. 5) that secures a certain free space in a block, a block may be unnecessarily spanned when data is added, and a cluster in which physical I / O frequently occurs at the time of access may occur. , Storage efficiency and access efficiency are reduced.

The reason is that setting of a free area suitable for each cluster is not taken into consideration in terms of storage efficiency and access efficiency due to the characteristics of data added to the cluster. For example, when a certain free space is secured in a block as shown in FIG. 6, if a large number of logically connected records are added, there is a problem in terms of storage efficiency and access efficiency.

An object of the present invention is to set a suitable free area for each cluster by designating a setting location and a ratio of a free area for each cluster according to the characteristics of data to be added to the cluster in the future. Another object of the present invention is to provide a data storage control system capable of improving storage efficiency and access efficiency.

[0011]

According to the present invention, a storage source database file in which a cluster, which is a collection of records having a logical relationship with the root record, is located in the vicinity of a storage area around the root record In the data storage control system having the initial storage control means for inputting the record from the storage device and initially storing the input record in the storage destination database file while securing the free space, the initial storage control means is specified in advance by the user. , Clusters are stored as free space setting locations for each cluster
Within a block to be executed or a continuous block after storing a cluster
Information and the free space cluster data capacity to be set
Based on the ratio for the case of securing a free space to the cluster storage block, each cluster to the end of the block
Bro If set the free space of the specified ratios, to ensure the free blocks contiguous after cluster stored, which calculates the amount of free space, which is calculated after the cluster stored in
Thus , a data storage control system characterized by setting an empty area for the number of packets is obtained.

[0012]

According to the storage control of the database, the user designates the setting location and ratio of the free space for each cluster according to the characteristics of the data to be added in advance. If it is determined by the information of the specified setting location that the cluster secures a certain free area in the block storing the record,
Store records while securing free space at the specified ratio in the block. If it is determined that the cluster secures continuous free blocks after storing the cluster,
The number of empty blocks to be secured is calculated based on the specified ratio, and the empty blocks are set after storing the cluster.

[0013]

Embodiments of the present invention will be described below in detail with reference to the drawings.

Referring to FIG. 1, a data storage control system according to an embodiment of the present invention includes a storage source database file 10 as a file to be processed, a storage destination database file 20, and an input from a storage source database file 10. And an initial storage control unit 30 for initially storing the created record in the storage destination database file 20 in a state where a free area is secured.

The initial storage control unit 30 includes a record input control means 31, a free area information file 32, a free area setting control means 33, a free space ratio storage area 34,
Record storage control means 35 and cluster capacity calculation means 3
6, empty block number calculating means 38, and empty block setting means 39.

The record input control means 31 inputs a record from the storage source database file 10. The free space information file 32 contains information on whether a free space is set in each block specified by the user (whether to secure a certain free space in a block storing a cluster or to secure a continuous free block after storing a cluster). ), And information on the ratio of the free area to the data capacity of the cluster. The free area setting control means 33 determines the free space ratio indicating the ratio of the free area in the block based on the information acquired from the free area information file 32, and controls the setting of the free area. The free space ratio storage area 34 stores the determined free space ratio. The record storage control means 35 stores the record in the storage destination database file 20. The cluster capacity calculation means 36 calculates the data capacity of the cluster (cluster capacity). The block size information 37 indicates the block size of the storage destination database file 20. The number-of-free-blocks calculating means 38 calculates the number of free blocks from the block size, the cluster capacity, and the ratio of the free area of the block size (free area ratio). The empty block setting means 39 sets an empty block.

The combination of the free area setting control means 33, the free space ratio storage area 34, and the record storage control means 35 secures a certain free area in a block for storing a record, based on information on a designated set location. When it is determined that the cluster is a cluster to perform, it functions as a means for storing a record while securing an empty area of a specified ratio in the block. The combination of the free area setting control means 33, the cluster capacity calculating means 36, the block size information 37, the free block number calculating means 38, and the free block setting means 39 is a combination of clusters for securing continuous free blocks after storing the cluster. If it is determined that the number of free blocks to be secured is calculated based on the specified ratio,
It works as a means to set an empty block after cluster storage.

FIG. 2 shows a formula for calculating the number of empty blocks by the number of empty blocks calculating means 38, and FIG.
(B) shows the data storage image according to the present invention and the data image of the free space information file 32, respectively.

Hereinafter, referring to FIG. 1, FIG. 2 and FIG.
The operation of the present embodiment will be described.

The record input control means 31 inputs a record from the storage source database file 10. When the input record is a root record, the free area setting control unit 33 specifies whether the data is a cluster data for which a free area should be reserved in a block or a cluster data for which a free block should be reserved after storing the cluster. Judgment is made from the information of the free space information file 32 thus obtained. In the case of a cluster in which a free area is to be secured in a block, the free area setting control means 33 acquires the ratio of the free area from the free area information file 32 and sets the free space ratio as the free space ratio indicating the ratio of the free area in the block. It is set in the rate storage area 34. In the case of a cluster for which an empty block should be secured after storing the cluster, the empty area setting control means 33 sets 0 (zero) in the free space ratio storage area 34 so as not to set an empty area in the block.

The record storage control means 35 determines whether all the records of the cluster input by the record input control means 31 have a free area set in the block based on the value set in the free space ratio storage area 34. Is determined and stored in the storage destination database file 20 while setting a free area if necessary.

When the free area setting control means 33 determines that the cluster is a cluster for which a free block should be secured after storing the cluster, the cluster capacity calculating means 36 calculates the data capacity of the cluster based on the input record. calculate. The number-of-free-blocks calculating means 38 acquires the block size of the storage destination database file 20 from the block size information 37, and calculates the data capacity of the cluster calculated by the cluster capacity calculating means 36 and the free area information file 32.
The number of blocks is calculated from the ratio of the free area acquired from the above using the free block number calculation formula shown in FIG. The empty block setting means 39 sets empty blocks for the number of blocks calculated by the empty block number calculating means 38 after storing the cluster.

FIG. 3A shows a data storage image according to the present invention, and FIG. 3B shows a data image of the free space information file 32. According to the data of the free space information file 32, the first, fourth, and fifth clusters C1, C4, and C5 to be stored are clusters that secure a certain free space in the block. The ratios are 25%, 25% and 33%, respectively. 2
Clusters C2 and C3 to be stored third and third
Is a cluster for securing a free area as a continuous free block after storing the cluster. The ratio of the free area to the data capacity of the cluster is 200% and 50%, respectively.
%. When the number of free blocks is calculated for the second and third clusters C2 and C3 by the free block number calculation formula shown in FIG. 2, the following formulas 1 and 2 are obtained, respectively.

[0024]

(Equation 1)

[0025]

(Equation 2) As is clear from the above, according to the present invention, FIG.
As shown in (a), a free area suitable for each cluster is set.

[0026]

As described above, according to the present invention,
By specifying the setting location and ratio of the free space in accordance with the characteristics of data to be added to the cluster in the future, the physical I / O is suppressed when adding data after initial data storage and when searching for additional data. Storage efficiency and access efficiency are improved. The reason is that, by designating the setting position and the ratio of the free space for each cluster from the outside, a suitable free space can be set for each cluster in accordance with the characteristics of data to be added to the cluster in the future.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a data storage control system according to one embodiment of the present invention.

FIG. 2 is a diagram showing a calculation formula when calculating the number of free blocks by a free block number calculation unit.

FIG. 3A is a diagram showing a data storage image according to the present invention, and FIG. 3B is a diagram showing a data image of a free space information file.

4A is a diagram illustrating a data storage image, and FIG. 4B is a diagram illustrating a record addition image in a data storage control method for securing a free area as a continuous free block after storing a cluster.

5A is a diagram showing a data storage image and FIG. 5B is a diagram showing a record addition image in a data storage control method for securing a free area in a cluster storage block.

FIG. 6 is a diagram illustrating an example in which a method of setting a free area is not suitable for characteristics of additional data.

[Description of Signs] 10 Storage source database file 20 Storage destination database file 30 Initial storage control unit 31 Record input control means 32 Free space information file 33 Free space setting control means 34 Free space ratio storage area 35 Record storage control means 36 Cluster capacity Calculation means 37 Block size information 38 Free block number calculation means 39 Free block setting means

Claims (3)

(57) [Claims] 1. A record is input from a storage source database file (10) centered on a root record, in which a cluster, which is a collection of records having a logical relationship with the root record, is arranged in the vicinity of a storage area. In a data storage control system having an initial storage control means (30) for initially storing an input record in a storage destination database file (20) while securing an empty area, the initial storage control means (30) The cluster is specified as a free space setting location for each cluster specified in advance.
In a block that stores
Information and whether to set the free space cluster data.
Based on the ratio of data capacity, to provide space cluster storage block, a class to the last block
If you want to set a free area of the ratio specified for each data and secure it as a continuous free block after storing the cluster,
A data storage control system comprising: calculating a size of a free area; and setting a free area for the number of blocks calculated after storing the cluster. 2. The means (32) wherein the initial storage control means (30) allows a user to specify a set location and a ratio of a free area for each cluster according to characteristics of data added in advance.
If it is determined from the information of the specified setting location that the cluster is to secure a certain free area in the block storing the record, the cluster is to secure a specified ratio of free area in the block. Means for storing records (33, 34, 35), and when it is determined that the cluster is to secure continuous empty blocks after storing the cluster, the number of empty blocks to be secured is determined based on the specified ratio. 2. The data storage control system according to claim 1, further comprising means (33, 36, 37, 38, 39) for calculating and setting an empty block after storing the cluster. 3. The record input control means (31) for inputting a record from the storage source database file designated by a user.
And indicates, for each cluster to which the record input by the record input control unit belongs, whether to secure a certain free area in the block storing the cluster or to secure it as a continuous free block after storing the cluster. A free area information file (32) in which the user specifies the setting area of the free area and the ratio of the free area to the data capacity of the cluster, and information on the setting place of the free area acquired from the free area information file. If it is determined that the cluster is a cluster that secures a certain free area in the block that stores the cluster, the free space rate indicating the ratio of the free area to be reserved in the block is used as the free space rate from the free space information file. The free space ratio to be acquired is set in the free space ratio storage area (34), and If the static is determined to be a cluster to ensure continuous free block after cluster storage, space area setting control means for setting 0 to the free space ratio storage area (34) (3
3) and the record input by the record input control means is
A record storage control means (35) for storing in the storage destination database file designated by a user while considering whether or not a free area is set in a block based on the value of the free space ratio storage area; If the setting control unit determines that the cluster is a cluster that secures a continuous free block after storing the cluster, a cluster capacity calculation unit (36) that calculates the data capacity of the cluster; Free block number calculating means (38) for calculating the number of free blocks to be secured from the block size information (37) of the storage destination database file, the ratio of free space obtained from the free space information file, and the data amount of the cluster; Means for calculating the number of free blocks after storage in the cluster (38)
2. The data storage control system according to claim 1, further comprising: an empty block setting means (39) for setting empty blocks corresponding to the number of empty blocks calculated in (1).