JP2658964B2 - Data record storage position control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data record storage position control method, and more particularly to a data record storage position in a data file in which data records are stored on a storage medium according to a certain order and data records are frequently updated and deleted. Related to the control method.

[0002]

2. Description of the Related Art In a conventional data record storage position control method, when data records are updated and deleted, and all data records in a block are deleted, an unnecessary area of the block is stored on a storage medium. The system was made reusable.

[0003]

In the above-described conventional data record storage position control method, in a data file in which update processing and deletion processing are repeated, a block in which a data record is arranged at the time of initial storage is one in the block. Even if some data records were deleted, the area was held as unnecessary free space in the block, so as a result of the update processing and deletion processing, the data record arrangement in the data file was biased, Unnecessary free space generated in the block due to the deletion process is not released until all the data records in the block are deleted, and there is a problem that it is difficult to reuse the unused space.

SUMMARY OF THE INVENTION In view of the above, it is an object of the present invention to store and manage a data record designated by a user so that an unnecessary empty area held on a storage medium can be reused by collecting and managing block by block. It is to provide a position control method.

The prior art is disclosed in Japanese Patent Application Laid-Open No. 5-250.
There are a data record storage method disclosed in Japanese Patent Application Laid-Open No. 234, No. 234, a data storage control method disclosed in Japanese Patent Application Laid-Open No. Hei 5-165686, and the like. It is not intended to collect and manage each block so that it can be reused.

[0006]

According to the data record storage position control method of the present invention, a storage area for storing a data record is dynamically secured in block units in a data file, and a next block pointer is stored in a block. By having
In a data record storage position control method for controlling a storage position of a data record for a data file in which the order relation between blocks is managed in the data file, blocks are stored in accordance with the order relation between blocks from the data file.
Searches for the lock, reads the block on the front side into the front block buffer as the previous block , and reads the block on the back side.
Data block search means for reading the block as a subsequent block into the subsequent block buffer, a previous block remaining capacity obtaining means for obtaining the previous block remaining capacity of the previous block in the previous block buffer, and a data obtained by the previous block remaining capacity obtaining means. After the previous block remaining capacity storage area for storing the previous block remaining capacity, the rear block used capacity obtaining means for obtaining the post-block used capacity of the post-block in the post-block buffer, and after the post-block used capacity obtaining means A block used capacity storage area for storing a block used capacity, and a previous block remaining capacity and a rear block used capacity obtained from the previous block remaining capacity storage area and the rear block used capacity storage area, and of the previous block in the previous block buffer. All of the subsequent block in the subsequent block buffer in the free area It is determined whether or not the data record can be stored. If the data record cannot be stored, the block size comparison means calls the data block search means. A data file updating unit that is called when all data records of the subsequent block in the buffer are determined to be storable and updates the data file; and in the data file updating unit, the data record of the subsequent block in the rear block buffer Record data transfer means for transferring the next block pointer of the previous block in the previous block buffer to the next block pointer of the next block in the rear block buffer. Next b Block pointer updating means, in the data file updating means, a block releasing means for releasing a subsequent block in a subsequent block buffer as a reusable empty block, and in the data file updating means, a preceding block and a following block in a preceding block buffer. Block overwriting means for overwriting a subsequent block in the block buffer with a corresponding block in the data file.

[0007]

[Action] In the data record storage position control method of the present invention, Bro data block retrieval means from the data file
Blocks are searched according to the order relationship between blocks, and
Block is read into the previous block buffer as the previous block.
Out look, it reads as a post-block block to be behind the rear block buffer, before block remaining amount acquiring means acquires the previous block remaining capacity before block before the block in the buffer, before the block remaining amount storage area previous block The previous block remaining capacity obtained by the remaining capacity obtaining means is stored, the rear block used capacity obtaining means obtains the post-block used capacity of the post-block in the post-block buffer, and the post-block used capacity storage area is the post-block used capacity. The post-block use capacity acquired by the acquisition means is stored, and the block capacity comparison means acquires the pre-block remaining capacity and the post-block use capacity from the pre-block remaining capacity storage area and the post-block use capacity storage area, and All of the subsequent block in the subsequent block buffer is stored in the free area of the previous block. Judge whether or not the data record of the previous block can be stored. If the data record cannot be stored, the data block retrieving means is called. Is called when it is determined that all the data records of the subsequent block can be stored, the data record transfer means transfers the data record of the subsequent block in the subsequent block buffer to the free space of the previous block in the previous block buffer, The next block pointer updating means rewrites the next block pointer of the previous block in the previous block buffer with the next block pointer of the next block in the next block buffer, and the block releasing means re-uses the next block in the rear block buffer. And the block overwriting means Overwrite block after the previous block and the succeeding block buffer in a block buffer to the appropriate block in the data file.

[0008]

Next, the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of a data record storage position control method according to an embodiment of the present invention. The data record storage position control method of the present embodiment includes a data file 1, a data block search unit 2, a previous block buffer 21, a rear block buffer 22, a previous block remaining capacity acquisition unit 3, and a previous block remaining capacity storage. Area 4, rear block used capacity obtaining means 5, rear block used capacity storage area 6, and block capacity comparing means 7
And a data file updating means 8 including a data record transfer means 9, a next block pointer updating means 10, a block releasing means 11 and a block overwriting means 12.

The data file 1 dynamically secures a storage area for storing data records in units of blocks internally, and has a next block pointer in a block,
A file in which the order relation between blocks is managed internally.

[0011] data block retrieval means 2, prior to manage the block buffer 21 and the rear block buffer 22,
Blocks from data file 1 according to the order relationship between blocks
Search for locks and replace the previous block with the previous block
Read out as the front block to the
The lock is read into the rear block buffer 22 as a rear block.
Protrude. More specifically, if the next block pointer of the subsequent block in the subsequent block buffer 22 points to an empty block , the data block search means 2 extracts the block indicated by the next block pointer of the previous block in the previous block buffer 21 from the data file 1 to the subsequent block. The data is read into the buffer 22, and if the next block pointer of the subsequent block in the subsequent block buffer 22 does not point to an empty block, the subsequent block buffer 2
After transferring the subsequent block in 2 to the previous block buffer 21, the block pointed to by the next block pointer of the previous block in the previous block buffer 21 is read from the data file 1 to the rear block buffer 22.

The previous block remaining capacity obtaining means 3 obtains the remaining capacity of the previous block in the previous block buffer 21 (hereinafter referred to as the previous block remaining capacity).

The previous block remaining capacity storage area 4 stores the previous block remaining capacity acquired by the previous block remaining capacity acquisition means 3.

The post-block used capacity obtaining means 5 obtains the used capacity of the post-block in the post-block buffer 22 (hereinafter referred to as the post-block used capacity).

The post-block used capacity storage area 6 stores the post-block used capacity obtained by the post-block used capacity obtaining means 5.

The block capacity comparing means 7 stores the previous block remaining capacity storage area 4 and the rear block used capacity storage area 6.
The remaining capacity of the previous block and the used capacity of the subsequent block are acquired, and it is determined whether or not all data records of the subsequent block in the subsequent block buffer 22 can be stored in the free area of the previous block in the previous block buffer 21, and cannot be stored. In this case, the data block search means 2 is called, and if the data can be stored, the data file updating means 8 is called.

The data file updating means 8 updates the data file 1 when it is determined from the comparison result of the block capacity comparing means 7 that the data can be stored.

The data record transfer means 9 stores the data record of the subsequent block in the rear block buffer 22 in the front block buffer 21 in the data file updating means 8.
To the empty area of the previous block in

The next block pointer updating means 10 is provided in the data file updating means 8 so that the previous block buffer 2
1 is rewritten with the next block pointer of the previous block in the rear block buffer 22.

The block releasing means 11 releases the subsequent block in the rear block buffer 22 as a reusable empty block in the data file updating means 8.

The block overwriting means 12 overwrites the corresponding block of the data file 1 with the preceding block in the preceding block buffer 21 and the succeeding block in the following block buffer 22 in the data file updating means 8.

Referring to FIG. 2, the processing of the data block search means 2 includes a head block reading step S21,
Next block pointer acquisition step for subsequent block S22
, Next block pointer end determination step S23, next block pointer empty block indication determination step S24
, A step S25 of transferring the subsequent block to the previous block buffer, a step S26 of acquiring the next block pointer of the previous block, and a step S27 of reading the next block pointer designated block.

Referring to FIG. 3, the processing of the block capacity comparing means 7 includes a previous block remaining capacity acquisition step S71.
And a subsequent block used capacity acquisition step S72 and a previous block remaining capacity / rear block used capacity comparison step S73.
And data block search means calling step S74;
It comprises a data file updating means calling step S75.

Referring to FIG. 4, the processing of the data file updating means 8 includes a data record transfer step S81,
It comprises a next block pointer updating step S82, a subsequent block releasing step S83, a data file overwriting step S84, and a data block searching means calling step S85.

FIGS. 5A and 5B illustrate the status of blocks in the data file 1 when updating the data file 1 based on the data record storage position control method of the present embodiment. FIG. Indicates the storage order of the data records, and the data record corresponding to and is deleted in the block of block number 01, and the data record corresponding to and in the block of block number 03. The used capacity of a block means the capacity from the first data record to the last data record, and includes a free area generated between data records. On the other hand, the remaining capacity of the block means the remaining capacity in the block after the last data record,
Do not include the space that occurs between data records.

Next, the operation of the data record storage position control method of the present embodiment configured as described above will be described.

First, the data block search means 2 reads the first block of the data file 1 designated by the user into the rear block buffer 22 (step S2).
1).

Next, the data block search means 2 acquires the next block pointer of the subsequent block in the subsequent block buffer 22 (step S22), and determines whether or not the next block pointer is a value LST indicating that there is no next block. Is determined (step S23).

If the next block pointer is not the value LST, the data block search means 2 determines whether or not the next block pointer has a value 0 indicating that the succeeding block is an empty block (step S24).

If the value of the next block pointer is not 0, the data block search means 2 transfers the subsequent block in the subsequent block buffer 22 to the previous block buffer 21 and sets it as the previous block (step S25).

Subsequently, the data block search means 2 acquires the next block pointer of the previous block in the previous block buffer 21 (step S26), and reads out the block pointed to by the acquired next block pointer to the subsequent block buffer 22 (step S26). S27).

The previous block is stored in the front block buffer 21 by the data block search means 2,
When each of the subsequent blocks is set in the second block 2, the previous block remaining capacity obtaining means 3 sets the previous block buffer 2
The remaining capacity of the previous block of the previous block in 1 is obtained and stored in the previous block remaining capacity storage area 4.

Further, the rear block used capacity obtaining means 4 comprises:
The post-block use capacity of the post-block in the post-block buffer 22 is acquired and stored in the post-block use capacity storage area 6.

Subsequently, the block capacity comparison means 7 obtains the previous block remaining capacity and the rear block used capacity from the previous block remaining capacity storage area 4 and the rear block used capacity storage area 6 (steps S71 and S72), and obtains the previous block. It is determined whether all data records of the subsequent block in the subsequent block buffer 22 can be stored in the free area of the previous block in the buffer 21 (step S73).

If all data records of the subsequent block in the subsequent block buffer 22 cannot be stored in the free area of the previous block in the previous block buffer 21, the block capacity comparing means 7 sets the data block searching means 2 to the entrance A (Step S74), and the next preceding block and the following block are stored in the preceding block buffer 21 and the following block buffer 22, respectively.

On the other hand, if all data records of the succeeding block in the succeeding block buffer 22 can be stored in the free area of the preceding block in the preceding block buffer 21, the block capacity comparing means 7 Called (step S75).

The data file updating means 8 firstly transmits data to the rear block buffer 22 by the data record transfer means 9.
All the data records of the subsequent block in the block are sequentially transferred to the free area of the previous block in the previous block buffer 21 (step S81). For example, the data record in the block with the block number 03 shown in FIG. 5A is transferred after the data record in the block with the block number 01, as shown in FIG. 5B.

Next, in the data file updating means 8, the next block pointer updating means 10 updates the next block pointer of the previous block in the previous block buffer 21 with the next block pointer of the subsequent block in the rear block buffer 22 ( Step S82). For example, the block number 02 which is the next block pointer in the block with the block number 03 shown in FIG. 5A is overwritten with the block number 03 which is the next block pointer in the block with the block number 01, and FIG. , The next block pointer in the block with the block number 01 is set to the block number 02.

Subsequently, the data file updating means 8 rewrites the next block pointer of the succeeding block in the succeeding block buffer 22 to the block number 0 indicating an empty block by the block releasing means 11, and It is made reusable (step S83). For example, FIG.
The block number 02 which is the next block pointer in the block with the block number 03 shown in FIG. 5A is set to the block number 0 as shown in FIG. 5B.

Next, the data file updating means 8 overwrites the corresponding block in the data file 1 with the previous block in the previous block buffer 21 and the subsequent block in the rear block buffer 22 by the block overwriting means 12. To update the data file 1 (step S84).

Finally, the data file updating means 8 calls the data block retrieving means 2 at the entrance A (step S85).

When called from the data file updating means 8, the data block searching means 2 acquires the next block pointer of the succeeding block in the following block buffer 22 (step S22). The next block pointer always has the value 0
(YES in step S24), the rear block (empty block) in the rear block buffer 22 is not transferred to the front block buffer 21 (step S2).
5), and obtains the next block pointer of the previous block in the previous block buffer 21 (step S2).
6), the block pointed to by the next block pointer is read from the data file 1 to the subsequent block buffer 22 (step S27). For example, block number 0 shown in FIG.
The block of block number 02 indicated by the next block pointer in the block of 1 is read out to the subsequent block buffer 22.

In this manner, the processing of the data block search means 2, the previous block remaining capacity acquisition means 3, the rear block use capacity acquisition means 5, the block capacity comparison means 7 and the data file update means 8 is repeated, and the data block search means In step S23 of step 2, the next block pointer is set to the value L.
When it comes to ST, the process ends.

In the above embodiment, the next block pointer is a block number. However, it is needless to say that the present invention can be similarly implemented by using another pointer such as a block head address.

[0045]

As described above, the present invention provides a data block search means, a previous block buffer, a rear block buffer, a previous block remaining capacity acquisition means, a previous block remaining capacity storage area, a rear block used capacity acquisition means, and a rear block. A used capacity storage area, a block capacity comparing means, a data file updating means, a data record transferring means, a next block pointer updating means, a block releasing means and a block overwriting means are provided. When the data can be stored in the empty area of the next block, all data records of the subsequent block are transferred to the empty area of the previous block, and the subsequent block can be reused as an empty block, so that the data records are managed for each block. Collect unnecessary free space in the data file There is an effect that it is possible to the main area reusable state on the storage medium.

Further, by controlling the arrangement of data records for each block, which is a unit of physical I / O, physical I / O at the time of accessing all data records in the data file is reduced, thereby improving access efficiency. Also has the advantage of being connected.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a data record storage position control method according to an embodiment of the present invention.

FIG. 2 is a flowchart showing a process of a data block search unit in FIG. 1;

FIG. 3 is a flowchart showing a process of a block capacity comparing means in FIG. 1;

FIG. 4 is a flowchart showing a process of a data file updating unit in FIG. 1;

FIG. 5A illustrates the state of a block in a data file before processing according to the data record storage position control method according to the embodiment, and FIG. 5B illustrates the state after processing according to the data record storage position control method according to the embodiment. FIG. 8 is a diagram illustrating an example of the status of blocks in the data file of FIG.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 data file 2 data block search means 3 previous block remaining capacity acquisition means 4 previous block remaining capacity storage area 5 rear block used capacity acquisition means 6 rear block used capacity storage area 7 block capacity comparison means 8 data file updating means 9 data record transfer Means 10 Next block pointer updating means 11 Block releasing means 12 Block overwriting means 21 Previous block buffer 22 Rear block buffer

Claims (3)

(57) [Claims] 1. A storage area for storing a data record is dynamically secured in block units in a data file.
By having a next block pointer in a block, in a data record storage position control method for controlling a storage position of a data record for a data file in which the order relation between blocks is managed in the data file, According to the order relationship between the blocks
A block is searched, the block on the front side is read out to the previous block buffer as the previous block , and the block on the rear side is read.
Data block search means for reading a block as a subsequent block into a subsequent block buffer, a previous block remaining capacity obtaining means for obtaining a previous block remaining capacity of a previous block in a previous block buffer, and a previous block remaining capacity obtaining means. The previous block remaining capacity storage area for storing the previous block remaining capacity, the rear block used capacity obtaining means for obtaining the post-block used capacity in the rear block in the rear block buffer, and the post-block used capacity obtaining means A post-block used capacity storage area for storing a post-block used capacity; a pre-block in a pre-block buffer obtained by acquiring a pre-block remaining capacity and a post-block used capacity from the pre-block remaining capacity storage area and the post-block used capacity storage area; To the free area of the rear block buffer. It is determined whether or not all data records can be stored. If storage is not possible, a block capacity comparison unit that calls the data block search unit; A data file updating unit that is called when all the data records of the subsequent block in the subsequent block buffer are determined to be storable and updates the data file; A data record transfer means for transferring a data record to a free area of a previous block in a previous block buffer; and the data file updating means for setting a next block pointer of the previous block in the previous block buffer to a position next to the succeeding block in the subsequent block buffer. Rewrite to block pointer Next block pointer updating means, in the data file updating means, a block releasing means for releasing a subsequent block in a subsequent block buffer as a reusable empty block; and in the data file updating means, A block overwriting means for overwriting a block and a subsequent block in the subsequent block buffer with a corresponding block in the data file. 2. The data block search means according to claim 1, wherein, if the next block pointer of the next block in the next block buffer points to an empty block, the data block search unit searches the next block pointer of the next block of the previous block in the previous block buffer from the data file. If the next block pointer of the subsequent block in the subsequent block buffer does not point to an empty block, the next block pointer of the previous block in the previous block buffer is transferred after transferring the subsequent block in the subsequent block buffer to the previous block buffer. 2. The data record storage position control method according to claim 1, wherein the block indicated by the data record is read from the data file to a subsequent block buffer. 3. The data record storage position control method according to claim 1, wherein said next block pointer is a block number.