JPH0887433A - File system block management system - Google Patents

Abstract

PURPOSE: To suppress the generation of an unnecessary storage area while speeding up input to and output from a disk storage device for a large-volume file. CONSTITUTION: A file system generating means 1 provides data block storage areas 102 and 103 for a disk storage device 101; and blocks b1-b5 of small size are generated in the data block storage area 102 and blocks b6-b10 of large size are generated in the data block storage area 103. A free block allocating means 2 when allocates necessary blocks so as to store file data inspects the kind of the file and decides which of the blocks of large size in the data block storage area 103 and the blocks of small size in the data block storage area 102 should be allocated.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a file system block management system using a disk storage device.

[0002]

2. Description of the Related Art The operation of a conventional file system block management system will be described with reference to FIGS.

FIG. 3 shows a file system in which a data block storage area 102 is allocated to a disk storage device 101 and blocks b1 to b10 having a fixed length of 4 Kbytes are created in the data block storage area 102.
Directory files F1 and 16 with a size of 3 Kbytes
FIG. 9 is a data configuration diagram when an image file F2 having a size of K bytes is created.

Data structures index1 and index2
Are the sizes of files F1 and F2, file types,
The block number of the block used by the file shall be stored.

Since the directory file F1 has a size of 3 Kbytes, one block of the block b1 is allocated from the data block storage area 102, and the data structure index1 stores the block number of the block b1.

Since the image file F2 has a size of 16 Kbytes, the block b is stored in the data block storage area 102.
4 blocks of 2, b3, b4, b5 are allocated,
The data structure index2 has blocks b2, b3, and b.
The block numbers 4 and b5 are stored.

Since blocks b1 to b5 are all 4 Kbytes, the total wasteful area generated at this time is block b.
It is 1K bytes included in 1.

Further, when reading all the data of the image file F2, the blocks b2, b3, b4, and b5 are read, so that the reading of 4 Kbyte unit is executed four times to the disk storage device 101.

FIG. 4 shows a file system in which a data block storage area 102 is allocated to the disk storage device 101 and blocks b1 to b5 having a fixed length of 16 Kbytes are created in the data block storage area 102.
Directory files F1 and 16 with a size of 3 Kbytes
FIG. 9 is a data configuration diagram when an image file F2 having a size of K bytes is created.

Since the directory file F1 has a size of 3 Kbytes, one block of the block b1 is allocated from the data block storage area 102, and the data structure index1 stores the block number of the block b1.

Since the image file F2 has a size of 16 Kbytes, the block b is stored in the data block storage area 102.
1 block of 2 is allocated, and the data structure ind
ex2 stores the block number of the block b2.

Since blocks b1 to b2 are all 16 Kbytes, the total wasteful area generated at this time is 13 Kbytes included in block b1.

Further, when reading all the data of the image file F2, the block b2 is read, so that reading in 16 Kbyte units to the disk storage device 101 is performed by one.
Will be executed once.

[0014]

In the conventional file system as described above, there is only one type of block size which is an access unit for performing input / output to / from the disk storage device.

For this reason, if the block size is increased, the useless area of the block storing the last part of the file becomes large, and the utilization rate of the disk storage device decreases. .

Further, if the block size is made small, when accessing data to a large capacity file, input / output to / from the disk storage device is frequently repeated in units of a small block size, resulting in a heavy load on the system. In addition, there is a problem that the input / output of a large capacity file to the disk storage device cannot be speeded up, that is, the input / output performance of the file cannot be improved.

In consideration of such problems of the conventional file system, the present invention can further improve the utilization rate of the disk storage device as compared with the conventional one and further improve the file input / output performance. An object of the present invention is to provide a block management system for a file system.

[0018]

According to a first aspect of the present invention, a disk storage device for storing a plurality of file data and a storage area of the disk storage device are divided to create a plurality of data block storage areas. File system creating means for creating an unused block group by dividing each of the data block storage areas into blocks of a different fixed length size, and allocating the blocks for storing the file data, Of the data block storage areas, data block storage area information indicating which data block storage area should be used is obtained, and free block allocation means for performing the allocation based on this information, and the file data are stored in the disk storage area. When inputting to and / or outputting from the disk storage device, , And file data input-output means for the input and / or output based on the unit of the size of the allocated block by block allocation means is a file system block management system with a.

According to the present invention of claim 2, file attribute information indicating the data size and / or the type of the file is attached to the data file, and the free block allocating means allocates the block. , Which of the data block storage areas should be used based on the file attribute information is provided, and obtaining the data block storage area information is performed using the determination criteria. It is a block management system of a file system which is to obtain data block storage area information.

[0020]

According to the present invention, the disk storage device stores a plurality of file data, the file system creating means divides the storage area of the disk storage device to create a plurality of data block storage areas, and each of the data blocks is stored. When the storage area is divided into blocks each having a fixed length size to create an unused block group, and the free block allocating unit allocates the block in which the file data is stored, among the plurality of data block storage areas , Data block storage area information indicating which data block storage area is to be used, the allocation is performed based on the information, and file data input / output means inputs the file data to the disk storage device. And / or the free block when outputting from the disk storage device. Wherein the input and / or performing the output on the basis of a unit of the size of the block allocated by blowing members Ri.

[0021]

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

FIG. 1 shows a block diagram of a block management system of a file system according to an embodiment of the present invention, and the configuration of the present embodiment will be described with reference to FIG.

In FIG. 1, 1 is a file system creating means, 2 is a free block allocating means, 3 is file data input / output means, 4 is a program for inputting / outputting files, 101 is a disk storage device, and 102 and 103 are disk storages. A data block storage area formed by dividing the storage area of the device 101, b1, b2, b3, b4, and b5 are blocks of the data block storage area 102, and b6 and b7.
And b8, b9, and b10 are data block storage areas 103
Is a block.

The file system creating means 1 allocates the data block storage areas 102 and 103 to the disk storage device 101 according to an instruction from the operator, and the blocks b1 to b5 having a fixed length of 4 Kbytes are allocated to the data block storage area 102. , Data block storage area 10
In 3, a file system including blocks b6 to b10 having a fixed length size of 16 Kbytes is created.

FIG. 2 shows a directory file F having a size of 3 Kbytes for the file system created in FIG.
FIG. 6 is a data configuration diagram when an image file F2 having a size of 1 and 16 Kbytes is created, and the operation of this embodiment will be described with reference to the diagram.

Here, the data structures index1 and in
It is assumed that dex2 stores the sizes of the files F1 and F2, the file type, and the block number of the block used by the file.

The file attribute information of the present invention is, for example, information about so-called file attributes such as file size and file type such as directory file or image file. In this embodiment, Focusing only on the type of the file, the case of using it as the file attribute information of the present invention will be described.

That is, the free block allocation means 2
Has a determination criterion that a data block storage area 102 is associated with a file whose file type is a directory, and a data block storage area 103 is associated with a file whose image is a file type.

Program 4 is directory file F1
Free block allocation means 2 when creating
Checks the type of the directory file F1 to identify it as a directory, and allocates one block of the unused block b1 from the corresponding data block storage area 102 according to the above-mentioned determination criteria, and the data structure inde
x1 stores the block number of the block b1.

When the program 4 accesses the directory file F1, the file data input / output unit 3 inspects the type of the directory file F1 and determines that the block size is 4 Kbytes. Input / output is performed in the storage device 101.

Next, when the program 4 creates the image file F2, the free block allocation means 2 inspects the type of the image file F2 and identifies it as an image,
According to the above determination criteria, one block of the block b6 is allocated from the corresponding data block storage area 103,
The data structure index2 stores the block number of the block b6.

Then, the program 5 sends the image file F2.
When accessing, the file data input / output unit 3 inspects the type of the image file F2, and the block size is 1
It is determined to be 6 Kbytes, and input / output is performed to the disk storage device 101 in units of 16 Kbytes.

As described above, according to the present embodiment, since the block of 4 Kbyte size is used for the directory file F1, the total wasteful area is 1 Kbyte size included in the block b1. Further, since a block of 16 Kbyte size is used for the image file F2, when reading all the data of the image file F2, in order to read the block b6, reading is performed in 16 Kbyte units to the disk storage device 101. It will be executed once.

As described above, according to the present embodiment, a plurality of data block areas each having a different block size are provided, and each file type is associated with which of the plurality of data block storage areas is to be used. By using an appropriate block size for each file type, it is possible to improve the utilization rate of the disk storage device.
It is also possible to improve the file input / output performance.

That is, since a data block storage area having a different block size can be designated for each file type,
For files with small size such as texts and directories, it is possible to reduce the size of the waste area generated by making the data block storage area of the small block size correspond to improve the utilization rate of the disk storage device. it can.

Further, for a file such as moving image data that is large in size and requires high-speed access, a data block storage area having a large block size is made to correspond to increase the access size unit to the disk storage device. I / O performance can be improved by reducing the number of accesses.

In this case, at first glance, it seems that the wasteful area generated in the data block storage area of this large block size increases, but this is not the case. That is, since only one block for storing the last part of the file per file causes a waste area, if one file size is large, there are many blocks without a waste area and one block with a waste area. With this configuration, the ratio of blocks in which a waste area is generated is reduced, and therefore the storage device utilization rate does not decrease.

In the above embodiment, the case where only two file types, that is, an image file and a directory file, are used as file types has been described.
Not limited to this, a text file may be added as another file type in the file type.

In the above embodiment, the file type is used as the file attribute information, and the case of using the file type is described. However, the present invention is not limited to this, and the file size is used as the file attribute information. Alternatively, the size may be used, and it is of course possible to use a criterion that the size of the file is equal to or larger than a predetermined value.

Further, in the above-mentioned embodiment, the case where the file type is focused on as the file attribute information and the block is allocated using this has been described, but the present invention is not limited to this, and for example, which data is explicitly specified from the user program. Information on whether to use the block area may be set, and the data block area to be used by the file may be determined from the setting information.

Further, in the above embodiment, when the program 4 accesses the directory file, the file data input / output unit 3 checks the type of the directory file and, for example, according to the check result that the block size is 4 Kbytes. The case has been described where the input / output is performed in the disk storage device 101 in units of 4 Kbytes.
Not limited to this, for example, when the free block allocating means 2 already creates the directory file described above, the unit of the block size of the block allocated in the corresponding data block storage area is directly allocated to the free block. It may be configured to be obtained from the means 2.

[0042]

As is apparent from the above description, the present invention can further improve the utilization rate of the disk storage device as compared with the prior art and further improve the file input / output performance. It has the advantage of being able to do it.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a block management system of a file system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a data configuration of a file system in the block management system of the file system of the present embodiment.

FIG. 3 is a diagram showing a data structure of a file system in a block management system of a conventional file system.

FIG. 4 is a diagram showing a data structure of a file system in a block management system of a conventional file system.

[Explanation of symbols]

 1 File System Creation Means 2 Free Block Allocation Means 3 File Data Input / Output Means 4 Programs 101 Disk Storage 102 Data Block Storage 103 103 Data Block Storage

Claims (2)

[Claims] 1. A disk storage device for storing a plurality of file data, a storage area of the disk storage device is divided to create a plurality of data block storage areas, and each of the data block storage areas is fixed separately. File system creating means for creating an unused block group by dividing the block into long-sized blocks, and when allocating the block in which the file data is stored, which data block storage area among the plurality of data block storage areas Free block allocating means for obtaining data block storage area information indicating whether to use, and inputting the file data to the disk storage device and / or from the disk storage device. When outputting, it is assigned by the free block assigning means. A block management system for a file system, comprising: file data input / output means for performing the input and / or output based on the unit of the size of the stored block. 2. The data file is provided with file attribute information indicating the size of the file and / or the type of the file, and the free block allocation means, when allocating the block, which of the data blocks Based on the file attribute information, a determination criterion for determining whether to use a storage area is provided, and obtaining the data block storage area information means using the determination criterion to determine the data block storage area information. The block management system for a file system according to claim 1, wherein the block management system is obtained.