JPH0283639A - File generation control method - Google Patents

Abstract

PURPOSE:To efficiently use a storage area by storing only data to be changed or filed data in another storage area in a disk at the time of updating the data, and performing access control on the same file data by a corresponding multiplex access managing table. CONSTITUTION:File managing structure is equipped with a read/write control (d) provided corresponding to each address storage area on a file address table, and the multiplex access managing table provided corresponding to each storage unit in a file data storage area in the disk. Since the sharing of the file data is performed only by updating information in a data multiplex managing area, it is not required to designate another storage area in the disk and to store the same file data in the area. When the data is changed, only the file data to be changed is stored in a new storage area in the disk according to the information in a change managing area, then, it is updated. In such a way, the storage area in the disk can efficiently be used.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a file creation control method in a device that stores significant data in the form of a file, and particularly to a file creation control method for creating new file data based on already stored file data. This article relates to a file creation control method for creating files.

[Prior Art] As a device for storing a large amount of data, there is a storage device that uses an optical disk, a magnetic disk, or the like as a storage medium. In such storage devices, meaningful data is stored on disk storage media in the form of files.

In such storage devices, significant data is stored in file units, and the file management method shown in Figure 5 is generally used to facilitate file searches and to store files regularly. A method is being adopted.

Referring to FIG. 5, the conventional file management structure shows a file name management table for file identification and a file storage area provided in a predetermined area of a storage medium such as an optical disk or a magnetic disk (hereinafter simply referred to as a disk). It includes a file address table and a storage area that actually stores file data.

The file name management table consists of a file name storage section a that stores a file name given to each file to identify the file, and a file address table pointer storage section that is provided corresponding to the file name storage section. Ru. The file name and file address table pointer are stored in a linked manner. The file address table pointer specifies the address of the file address table.

The file address table is addressed according to the file address table pointer contained in the file name management table. This file address table is provided for each file, and stores a disk (block) address indicating a storage location on the disk where data of the corresponding file is stored.

The file data storage area stores data in blocks, and each block is assigned a block address C indicating an address in the storage area. Next, file search and read operations will be briefly explained.

When searching for a file, a file name for file identification is first given to the storage device. In the tα device, an area for storing a file name corresponding to the given file name is searched in the file name management table. In this file name management table, when a file name storage part a that matches a given file name is found, the file address table is accessed according to the file address table pointer b linked to the found file name storage part. Ru.

The file address table stores block addresses indicating locations where requested file data is stored. Therefore, the storage area is accessed according to the block address stored in this file address table, and the file data stored at the block address is read out.

When writing a file, a file name is given and then a data write instruction is given. On the storage device side, in response to a data write and file storage instruction, a newly given file name is stored in the file name storage section a, and a new file address table pointer is set by linking to this file name. . A new file address table is set corresponding to the file address table pointer b, and the block address where the file data to be written is stored is stored in this file address table. Data writing is performed to the addressed block of the disk according to this block address.

[Problems to be Solved by the Invention] As mentioned above, in conventional disks, data to be saved is given a name as a file, and there is a difference between the file specified by the file name and the file actually stored in the storage area. There is a one-to-one correspondence with the entity of file data to be stored, and the file management is simple. However, when file data already stored on a disk is copied and used as new file data, the following problems occur.

Now, with reference to FIG. 5, consider the case where file B is created from file A. File A has a file address table pointer XX, and block addresses 3, 5, and 6.8 are stored in the file address table specified by this file address table pointer XX.

File data A is in block address 3 on the disk.
is stored, and file data B is stored at block address 5.
is stored, and file data C is stored at block address 6.
is stored, and file data D is stored at block address 8. When copying this file A and changing some of its data to create file B, the following operations are performed. First, a file B and a file address table pointer YY linked to this file B are set in the file name management table by a copy instruction signal. A file address table is set using this newly set address table pointer YY.

This file address table contains block addresses 7.9.10 and 1, which represent different storage areas within the disk.
1 is stored. This block address 7.9,10.
File data A, B, C, and D corresponding to file A are respectively stored in the storage areas of the disk corresponding to 11. After this, read this file B,
Required changes are made to the data stored in block addresses 7, 9, 10 and 11 and stored as new file data A'B'C' and D'.

As another method, file data A of file A,
After all files B, C, and D are read onto the main memory and necessary data changes are made therein, a procedure may be performed to store the changed file data as file B in the disk again. Even in this case, the storage areas of the file data that are finally driven into the disk are different for the original files A and B.

As mentioned above, in the conventional data management structure, when data that is already stored on the disk is used as data for creating a new file, the file data stored in the storage area of the disk is The entire entity of the file is transferred to another storage area, and then only necessary data changes are made to the copied file data, or - the basic file data is read out onto the main memory. After that, necessary data changes are made there, and then the data is saved in a separate storage area on the disk.

Therefore, even if the change is actually just a small part of the data (i.e., file data A.

A', B, B', C, and C' are completely the same and only the file data D and D' are different), the disk storage area is an area that stores the same file data. There may be a case where a plurality of such files exist, and the storage area is wasted. In particular, when the size of the file is large, the disk has a large number of storage areas for storing the same file data, causing a problem that the disk is using a lot of wasted storage area. In addition, with this file data creation method, if the file size is large, all of the underlying file data must be moved to another storage area, which increases the time required to copy the file. There was also the problem of slow file processing speed.

As mentioned above, in the conventional data management structure, the data stored on the disk is usually named as a file, and the file with this file name and the data stored in the storage area are There is a one-to-one correspondence with the actual entity, which simplifies file management.However, as mentioned above, when creating new file data based on data that has already been saved, storage Because it is necessary to create an area in the new area to store the same file data as the original saved data,
There was a problem that the tα area was used unnecessarily and the time required for this copying became longer.

Therefore, the first object of the present invention is to eliminate the drawbacks of the conventional data creation control method as described above, to make it possible to use the storage area of the disk efficiently, and to create data in a short time. The object of the present invention is to provide a data creation control method that allows the data creation to be controlled.

[Means for Solving the Problems] A file creation control method according to the present invention includes a table for managing which part of file data is multiplexed and shared by multiple files, and a table for managing which part of file data is multiplexed and shared by multiple files, and a table for managing which part of file data is multiplexed and shared by multiple files. Two management tables are provided: a change management table that can notify whether the file data part is requested to be changed and whether changes can be made to the block, and data is created by a combination of these two management tables. This is how it was done. Note that the term "change" mentioned above can be extended to the term "write".

That is, the data creation control method according to the present invention links and adds read/write controls indicating whether data specified by the address can be read or changed corresponding to each file address stored in the file address table. a step of linking and adding information indicating the number of files sharing the data to each area storing file data; and a step of linking and adding information indicating the number of files sharing the data to each area storing the file data, based on the added read/write control and the information on the number of shared files. and creating a new file.

When copying a file, there is a step to set the file address storage area for the newly created file, and a read/write control provided for the file address of the first saved file that is the basis. step, and a step of transferring all the file address values of the first file that is already saved as the basis to the newly set file address storage area, and a step of transferring the file address values of the first file that is already saved as the basis, and the step of incrementing the number information of co-intoxication files provided accordingly by one.

When updating data, the read/write control of the file to be updated is searched, and if the searched read/write control indicates only readable, the shared file set for the corresponding file data is searched. If the searched shared file number information is 1, the read/write control provided at the corresponding file address is changed to both readable and change (write) enabled states. , the step of accessing the corresponding file data and changing it to the desired data, and if the above-mentioned searched file information indicates 2 or more, decrementing the shared file number information by 1 and creating a new file data storage area. is set in the disk, and the address to be updated is replaced in the above file address table by the address of the newly set file data storage area, and the read/write control corresponding to the replaced file address is read. This includes the step of setting the file data storage area to a enabled and writable state, and then writing (changing) desired data to the newly set file data storage area.

[Operation] In this invention, file data is shared only by updating information in the data multiplex management area, so there is no need to set up a separate storage area in the disk and store the same file data there. do not have.

When changing data, only the file data to be changed is stored and updated in a new storage area within the disk according to the information in the change management area.

[Embodiment of the Invention] FIG. 1 is a diagram showing a file creation control method as an embodiment of the invention, and shows an example of a file management structure for two files.

Referring to FIG. 1, the file management structure used in the file control method which is an embodiment of the present invention includes a read/write control d provided corresponding to each address storage area of the file address table, and a A multiple access management table is provided corresponding to storage +1 (one block in the figure) of the file data storage area.

The read/write control indicates whether the file data indicating the corresponding block address is readable or writable. In the case of information "R", it indicates that the data of the corresponding block address can be read, and "W"
In the case of , the data of the corresponding block address is written (
change) is possible.

The multiple access management table is provided in a manner linked to each block address, and indicates how many files share the file data of this linked block address. In other words, if the file data of the corresponding block address is not used in any file, it is "0", if it is used only in Ill German files, it is 1, and if it is used in multiple files, it is "0". is set to the value of the number of shared files (2 to n).The operation will be explained next.

Now, let us consider the case where the file data of "File A" is copied to create "File B".

First, when an instruction is given from the outside to copy file "File A" and create a new file "File B," a name "File B" to specify the new file to be created is displayed in the file name management table. ” is stored. In parallel with this, a file address table "YY" for storing the disk address of this newly created file "File B" is set. The pointer for identifying this newly set file address table is file address table pointer b.
The new file "File B" is linked and stored in the area. Next, file address tables XX for both “files A and B”.

Read/write control added to YY,
Set to "R" indicating that all can be read.

Next, all the disk addresses of file address table XX of "File A" are changed to file address table Y.
Move to the address storage location of Y. That is, in the case of FIG. 1, block (disk) addresses 3, 5, and 6.8 of the file address table XX are stored in the address storage positions of the file address table YY, respectively. Next, the shared file number information (referred to as a reference counter in the following explanation for convenience of explanation) of the corresponding disk address, that is, block address 3, 5, 6.8 in the multi-m access management table, is incremented by one. As a result, the data at block addresses 3, 5, and 6.8 are shared between the two files "File A# and File B." By the above operation, "File A° is copied and " The operation of creating file B'' is completed.

Next, we will explain the control operation when updating (writing) either the file A# or the file B. Now, referring to FIG. The case of updating will be explained. In this case, first check the read/write controls added to the file address table YY linked to "File B". All the read/write controls added to the disk addresses stored in the file address table YY are ''R'' and all the data in "File B" is readable, then refer to the multiple access management table provided corresponding to the disk address stored in the file address table. The counter value can be checked. If this reference counter value is already 1'', then the data stored at that disk address is one file (in this case, “
Since it is only used for "File B") and can be updated without considering the effect on other files, the read/write control attached to the file address table YY can be changed from "R" to "R". It is changed to "RW". As a result, the contents of the disk address are updated to 1-07, so the data of this disk address is updated.

If the current read/write control is "R" and the reference counter value of the corresponding multiple access management table is any value from 2 to n, the data is stored at that disk address. This indicates that the data contained in the file is shared in several files. For this reason,
In this case, the value of the reference counter is decremented by 1, and a storage area for newly storing data at this address is acquired in the disk. This newly acquired block address of the storage area in the disk is linked to the file address table YY, and the read/write control added thereto is set to "RW".

The original file data is stored in this newly acquired storage nfi area.

At this time, the reference counter of the multiple access management table provided corresponding to the disk address indicating the storage address of the entity of the newly acquired file in the disk is set to 1''. Since the read/write control is set to "RW", access is performed to the newly changed disk address by χ.
File data is updated. In the configuration of FIG. 1, an example is shown in which block addresses 3.5 of file address table YY are changed to 9 and 10, respectively, and changed to file data C'' and D''. In this case, the newly acquired block address a1 (in-file storage area) is disk address 9 and 1.
It is address 0, and block addresses 9 and 10 are stored in corresponding areas of the file address table YY. Therefore, disk addresses 6 and 8 are shared by "File A" and "File B", and in "File B", file data A and B of "File A" are changed, and file data C1
1, the file data has been changed to D'. At this time,
Since the file data C1D is shared by "File A" and "File B'," the value of the corresponding reference counter in the multiple access management table is 2, and on the other hand, the value of the corresponding reference counter of the disk address C is 3. 5. 9. 10 are only shared in one file, and the value of the reference counter thereof is set to 1#. The above update operation can be summarized as follows. When the read/write control is "RW", the value of the reference counter in the multiple access management table is not referenced, and the file data at the corresponding disk address is updated as is.

If the read/write control is R'', the value of the corresponding reference counter in the multiple access management table is searched,
If its value is 1", the read/write control's
Rm is changed to "RW", thereby making it possible to update the file data at each disk address, and thereafter accessing that disk address to update the file data.

If the read/write control is "R", the value of the reference counter in the multiple access management table is searched and the value is shared by multiple files.
In the case of one shift from 2 to n'', the value of the corresponding reference counter is first decremented by 1.Next, a new storage area is acquired in the disk, and the acquisition is completed. The disk address indicating the storage area location is linked to the file address table.The read/write control provided corresponding to the linked disk space address is set to "RW" in 117, and the new The reference counter of the multiple access management table for the storage area acquired in “
1″. After this, that newly acquired memory,
Access the area and update file data.

Next, from the source file “File A”, create a new “
The operation when creating "File B" and "File C" will be specifically explained.

First, the source file “File A” that is already stored
The size of the disk is 3 bytes, that is, 3 disk blocks (IK bytes/block).

Now, in the configuration shown in FIG. 2, file data stored in each block is illustrated as A, B, and C.

Next, in accordance with an external copy command, the contents of "file A" are copied to create "file B." The operation at this time, that is, the operation when giving a copy instruction, may be the same as the conventional method, or may be different from the conventional method.

However, when using the conventional method and the 5° 4t method 1.2, it is necessary to configure the system so that the instruction is transmitted to the file control side.

On the file control side, in response to the copy command, "File B" is added to the file name management table, and
File address table YY to be linked to it
Create this file address table YY
Copy the contents of file address table Xx to.

Similarly, the value of the corresponding reference counter (in FIG. 2, the reference counter provided at disk addresses 3, 4, and 5) of the multiple access management table is incremented by one. During this copy operation, all read/write control data provided corresponding to the file address table is set to "R", and the file entity of the disk address stored in the file address table xx, yy is set to "R". Make any of them readable. This operation causes the file "File A" to be changed from "File B" to "File B".
The operation of creating the file ends.

When creating a new file "File C", create the file entity on the disk separately from the already stored "File A" and "File B" using the normal conventional method and 1111 method. By writing file data, it may be possible to link "File C" in the file name management table with the file entity on the disk. At this time, the value of the reference counter of the corresponding multiple access management table is set to "1".

However, if "File C" is created by copying from "File B", the above "File B"
This is the same as creating "file A" from "file A", and the same is true when creating "file C" from "file A" by copying. In other words, in the former case “
The relationship between file B'' and file C'' is the latter, and the latter is the relationship between file A and file C. In either of these relationships, the file address tables Xx, y
The disk addresses stored in y and zz are all the same addresses 3, 4.5, and the read/write controls added to them are all "R".

Also, at this time, the value of the reference counter in the multiple access management table provided for disk addresses 3, 4.5 becomes "3", indicating that this disk address is shared by three files.

Next, the operation when partially updating the contents of "file B" will be explained with reference to FIG. Now, assume that the file data to be updated is the file data corresponding to disk address 4, ie "Bo." Normally, when reading/writing the contents of a file, a file open command given from the outside is used to open the file. Use a seek command to move the pointer to the required position for the given file name.In other words, move the pointer to the point that matches the file address table pointer set up corresponding to this file name, and read the data from there. A desired file is read or written by reading/writing the size (3 blocks in the figure).Therefore, in this embodiment as well, reading/writing is performed in the same way as before, and now the "file" is read/written. If an update command is issued for the second relative block of "B", that is, disk address 4, and a seek is made to the location of the storage area corresponding to this disk address 4, the file system stores that location (pointer). (This stored pointer is stored in another register, for example).

The operation after a read command is issued in this state is the same as the conventional operation, and the contents of the sought disk address 4 are simply read.

When seeking to this disk address 4, if a write command indicating a change is issued, the file system first refers to the corresponding file address table read/write control, and if it is already "RW", The write (
or change) the operation ends. Now, if the read/write control added to the relative second block address of file address table YY is "R", it indicates that the contents of that disk address are shared by multiple files. Therefore, the reference counter of the multiplex management table provided for each disk address is referred to, and if the counter value is 1'', that block is only used by one file, so the read cannot be performed. / Change R" in the write control to "RW" indicating writable, access the file data block of the corresponding disk address, change the file entity in the disk to the desired data, and write the data ( or change) the operation ends.

In this case, if the final count value of the multiplex management table is 2
In the above case, the data storage area is decremented by 1, a new data storage area is acquired, a new disk file entity is stored there, and the disk address is set at the corresponding position in the file address table YY (i.e., the link )do. This operation is performed according to the stored pointer. Also, the reference counter value of the multiple access management table provided corresponding to the newly set disk address is initialized to 1". After this, the newly acquired file entity in the disk is accessed. Re,
The data modification is completed by making the desired data modification and storing it again at that location. In the configuration shown in FIG. 3, the file data "Bo" is at disk address 9.
A case is shown in which the file data is stored as "B" whose address has been changed. Here, the operation of changing the file entity within the disk includes an operation of reading the actual file entity within the disk, performing desired data changes, and then storing it again in the corresponding storage area (disk address).

The regularity of the above-mentioned operations, that is, the search for read/write control added to the file address table, the search for the counter value of the reference counter provided in the multiple access management table, and the rules regarding the operations performed in conjunction therewith are as follows: When updating a file, a new file is formed by branching from the source file that is the basis of the updated file, and is linked to the newly added file entity. Therefore, it is easily possible to generate an almost infinite number of copies of files, and even if this nearly infinite number of copies of files occurs and an update is instructed to any of the generated files, the above sequence You can easily respond according to the following.

When deleting an unnecessary file, as shown in Figure 4, simply delete the file name to be deleted from among the file names stored in the file name management table and the file address table pointer attached to it. Refer to the disk address stored in the file address table provided for the file to be processed, and refer to the multiple access management table provided for this disk address. ((This can be done by simply decrementing the counter value by 1. At this time, if the count value of the reference counter in the multiple access management table is decremented by 1 and becomes "0", then All that is required is to release the area of the corresponding disk address.FIG. 4 shows a state in which "file A" has been deleted from the file storage state shown in FIG.

[Effect of the invention] As described above, according to this invention, file creation 1;
As a control method, for the file address table provided for each file, the disk address (block address) stored in the corresponding file address table is
In addition to providing read/write control to control read/write (change) of the contents of
A multiple access management table is provided for each disk address indicating the multiplicity of the disk address, and operations such as data copying and updating are performed based on these read/write controls and the information stored in the multiple access management table. It is configured as follows. Therefore, during a file copy operation, in order to complete the copying of the file address table and incrementing the value of the B reference counter provided for the corresponding disk address in the multiple access management table by 1, the file entity is transferred to the disk. There is no need to transfer the data to another storage area, making it possible to perform copy processing at high speed. Particularly, when the size of the file to be copied is large, the effect becomes more pronounced as the file size increases.

In addition, when updating data, only the data to be changed and the data that has been filed is stored in a separate storage area on the disk, and access to the same file data is controlled by the corresponding multiple access management table. , the same file data is no longer stored in multiple storage areas, and storage areas can be used efficiently.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a file creation control method according to an embodiment of the present invention, and is a diagram showing a management structure for two files. FIG. 2 is a diagram showing a file creation control method in another embodiment of the present invention, and is a diagram showing a file structure when three files are created by copying from a source file. FIG. 3 is a diagram showing a file changing method in a file control method according to yet another embodiment of the present invention, and is a diagram showing a file structure when file data at a data address of one file is changed. FIG. 4 is a diagram showing the file deletion operation in the file creation control method according to the present invention, and is a diagram showing the file structure after one file is deleted. FIG. 5 is a diagram showing a file structure in a conventional file creation method. In the figure, a is a file name storage section, b is a file address table pointer storage section, C is a disk address (block address), and d is a read/write control. In addition, in the figures, the same reference numerals indicate the same or corresponding parts. (2 others)

Claims (1)

[Scope of Claims] (1) A file creation control method in a device that stores significant data in the form of a file, wherein the storage device includes a file name for specifying a file and a file linked to the file name. a file name management area that stores an address pointer, a file address area that is addressed by the pointer value of the file address pointer and stores an address in the storage device of the corresponding file data, and an area that stores the file data. and linking each of the file addresses stored in the file address area with read control and write control information indicating whether data specified by the address is writable or changeable, and each of the file data storage areas. a file creation control comprising: linking information on the number of files that share the data; and creating a new file based on the linked read control and write control and the shared file number information. Method. (2) The step of creating a new file includes creating a new file based on already stored first file data, and setting a file address storage area for the file to be newly created. a step of setting all read controls and write controls provided corresponding to the file address of the first file to a readable state; and setting all file address values of the first file to the newly set file address. 2. The file creation control method according to claim 1, comprising the steps of: copying to a storage area; and incrementing shared file number information provided corresponding to the file data of the first file by one. (3) The step of creating a new file includes a step of updating the file, and the step of updating the file includes searching for a read control and a write control of the file to be updated, and reading out the read control and write control that are found. a step of searching for information on the number of shared files provided with corresponding file data if only "possible" is indicated; and a step of searching for information on the number of shared files provided with corresponding file data; changing the state to both readable and writable, accessing the corresponding file data, and changing it to desired data; and when the retrieved shared file number information indicates 2 or more, the retrieved Decrement the shared file number information by 1, set a new file data storage area, and
The address indicating the file data to be updated in the file address storage area is replaced by the address of the newly set file data storage area, and read control and write control provided corresponding to the replaced file address are readable and The file creation control according to claim 1 or 2, comprising the step of setting both the writable state and writing desired data to the set new file data storage area. Method.