JPH04350738A - Disk space management system - Google Patents

Abstract

PURPOSE:To improve the use efficiency of the space on a magnetic disk device in a file system where a unit of input/output and area reservation is extended to plual blocks. CONSTITUTION:The reservation size of an area required to write data in unit logical areas L0 to L5 of the 32-block size, which constitute a logical area L of a file recognized by a user program, into a physical area P on the magnetic disk device is selected from three kinds of block number 1, 8, and 32 in accordance with the data volume to be written. That is, when new write of data having 22-block, 7-block, One-block, 3-block, one-block, and 29-block sizes on unit logical areas L0 to L5 is requested, 32 blocks of a unit physical area P0, 8 first blocks of a unit physical area P1, one first block of a unit physical area P2, 9th to 16th blocks of the unit physical area P1, the second block of the unit physical area P2, and 32 blocks of a unit physical area P3 are reserved in order, and data in unit logical areas L0 to L5 are written in these reserved blocks, and correspondence relations between logical areas and physical areas are independently managed.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention secures a file area using a continuous area of a certain size as a unit of input/output for the purpose of high-speed input/output of large-capacity files on a magnetic disk device. The present invention relates to file systems in electronic computers, and particularly to a disk space management method that can improve the efficiency of using physical areas on magnetic disk devices.

0002

[Background Art] A type of file system in electronic computers is used to define the logical area of a file recognized by a user program in units having a size that is an integral multiple (for example, 32 times) of one block, which is the input/output unit of a general file system. In addition to dividing the physical area into logical areas, the physical area on the magnetic disk device is divided into continuous unit physical areas of the same size, and the unit logical areas and unit physical areas are made to correspond one-to-one.
A file system has been reported that enables high-speed input/output of large-capacity files on a magnetic disk device by using a unit physical area as a unit of input/output.

0003

[Problems to be Solved by the Invention] In the above-mentioned file system, since the unit logical area and the unit physical area have a one-to-one correspondence, when writing data in a certain unit logical area to a magnetic disk device, , even if the size of the write data is much smaller than the size of the unit physical area.
One unit physical area will be secured.

For this reason, there has been a problem in that the space efficiency on the magnetic disk device is degraded.

The present invention solves these conventional problems, and its purpose is to expand the input/output unit to multiple blocks so that large amounts of file data can be input/output at high speed. The purpose of this invention is to improve the efficiency of using physical areas on magnetic disk devices in file systems for electronic computers.

[0006]

[Means for Solving the Problems] In order to achieve the above object, the present invention divides the logical area of a file recognized by a user program into unit logical areas of multiple block sizes, and also divides the physical area on a magnetic disk device. In the file system for an electronic computer, which speeds up the input and output of large-capacity file data by dividing the unit into continuous unit physical areas of the same size as the unit logical area and expanding the input/output unit into multiple blocks. a logical-physical mapping management table that maintains the correspondence between logical areas and the physical areas; a free area management table for managing unused physical areas on the magnetic disk device; and the logical-physical mapping. Managing the correspondence between the logical area and the physical area using a management table, managing the usage status of the physical area using the free area management table, and specifying a logical area number from the user program. When a new write request is made for the requested data, a secured area size suitable for the requested data size is determined from among multiple types of secured area sizes set to the same size as the unit physical area and several smaller sizes. Then, a reallocation processing means that secures a physical area of this determined secured area size from an unused unit physical area or a partially used unit physical area with the same secured area size, and and input/output control means for writing data of a unit logical area corresponding to the designated logical area number into the physical area through a magnetic disk device control means.

[0007]Furthermore, when the user program requests overwriting or additional writing of data specifying a logical area number, the reallocation processing means changes the data size after overwriting or additional writing to the data before being overwritten or additionally written. Check whether the size of the physical area that has already been reserved for A physical area of a size suitable for the data size is secured from an unused unit physical area or a partially used unit physical area with the same secured area size, and when additional writing is performed, the original data is transferred to the newly allocated physical area. The input/output control means is notified of the newly secured physical area by transfer, and the original secured area is released, and the input/output control means requests the overwriting or additional writing in the notified physical area. The data is written through the magnetic disk device control means.

[0008]

[Operation] In the disk space management method of the present invention, when a new write request for data specifying a logical area number is made from a user program, the reallocation processing means selects a unit physical area of the same size or several smaller sizes. The secured area size suitable for the requested data size is determined from among the multiple types of secured area sizes set in , and the physical area of the determined secured area size is used as an unused unit physical area or the same secured area size. The input/output control means writes the data of the unit logical area corresponding to the specified logical area number into the secured physical area through the magnetic disk device control means. .

[0009] Furthermore, when a user program requests overwriting or additional writing of data specifying a logical area number,
The reallocation processing means checks whether the data size after overwriting or additional writing exceeds the size of the physical area already reserved for the data before being overwritten or additionally written, and if it does not exceed the size, the data size is already reserved. The input/output control means writes the data requested to be overwritten or additionally written to the notified physical area through the magnetic disk device control means. On the other hand, if the data size is exceeded, the reallocation processing means transfers a physical area of a size suitable for the data size after overwriting or additional writing to an unused unit physical area or a partially used unit physical area with the same secured area size. At the time of additional writing, the original data is transferred to the newly secured physical area, the newly secured physical area is notified to the input/output control means, and the input/output control means transfers the data to the newly secured physical area. The data requested to be overwritten or additionally written is written through the magnetic disk device control means. In this case, the original reserved area in which data was recorded before being overwritten or additionally written is released by the reallocation processing means.

0010

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

Referring to FIG. 1, one embodiment of the present invention includes:
A high-speed file management means 1 used to speed up the input/output of large-capacity file data by expanding the allocation unit of the file data area, a logical area recognized by the user program 9, and a magnetic disk device 8. an input/output control means 2 that manages input/output in correspondence with the physical area of the magnetic disk device 8; a magnetic disk device control means 7 that controls reading from or writing to the magnetic disk device 8; A free area management table 3 for managing unused areas on the device 8, a logical-physical mapping management table 4 for maintaining the correspondence between logical areas and physical areas, and a magnetic disk device 8.
Transfer processing means 5 for changing the area where data is placed, and reallocation processing for improving the efficiency of disk space by using the above free area management table 3, logical/physical mapping management table 4 and transfer processing means 5. means 6.

FIG. 2 shows a logical area L of a file recognized by the user program 9 and a physical area P on the magnetic disk device 8.
It is a figure which shows the example of a structure with and the relationship between both. As shown in the figure, the logical area L recognized by the user program 9 is divided into unit logical areas L0 to L5 of 32 block size, and the physical area P on the magnetic disk device 8 is also a continuous unit physical area of the same 32 block size. It is divided into regions P0 to P5.

[0013] In this embodiment, one block is set as the reserved size of the area necessary for newly writing data on the unit logical areas L0 to L5 of 32 block size constituting the logical area L to the physical area P. A total of three types of sizes are set: a size for 1 minute, a size for 8 blocks, and a size for 32 blocks, and an appropriate secured area size is determined from among them according to the amount of data to be written.

That is, if the size of the data on the logical area for which writing is requested is 1 block, a physical area of 1 block size is secured, and if it is 2 to 8 blocks, a physical area of 8 blocks size is secured. If it is 9 to 32 blocks, a physical area of 32 blocks in size is secured.

FIG. 3 is a flowchart showing an example of the processing executed by the reallocation processing means 6 when a write request is received from the user program 9. The write operation of this embodiment configured as described above will be explained below. explain.

Now, the user program 9 is executing an input request to newly write 22 block size data (dotted area) in the unit logical area L0 of the logical area L of the input/output file as shown in FIG. When an output request is issued, this input/output request is transmitted to the input/output control means 2 through the high-speed file management means 1, and the input/output control means 2 passes control along with necessary information to the reallocation processing means 6.

Since the reallocation processing means 6 is writing new data, it proceeds to processing S2 via processing S1 in FIG. 2, and determines a size of 32 blocks as the secured area size corresponding to the current data size (22 blocks). .

Next, the reallocation processing means 6 reserves the above-determined physical area of 32 block size from the magnetic disk device 8 as an area to be used for writing the current data (S3).

Now, assuming that all unit physical areas P0 to P5 of the physical areas on the magnetic disk device 8 are managed as free areas in the free area management table 3, the reallocation processing means 6, for example, P0 no 3
Secure two blocks as the area to be used for writing the current data.

At this time, the reallocation processing means 6
In response to securing the unit physical area P0, the number of the unit logical area L0 (also referred to as the logical area number), the number of the unit physical area P0 (also referred to as the physical area number), and its block position ( 0 to 31), and since 32 blocks of the unit physical area P0 were used for allocation, the contents of the free area management table 3 are updated accordingly. These processes are performed in process S4.

[0021] When the above processing is completed, the reallocation processing means 6 sends the physical area information including the number of the unit physical area P0 secured in step S3 and its block position (0 to 31) to the input/output control means 2. Return it (S5).

The input/output control means 2, which has been notified of the physical area information, issues an input/output request to the block positions (0 to 31) of the unit physical area P0 to the magnetic disk device control means 7. In response, the magnetic disk device control means 7 receives the new write request from the user program 9.
Two block size data is written to the unit physical area P0 in the physical area P of the magnetic disk device 8.

In this way, as shown in FIG. 2, the 22 block size data of the unit logical area L0 is written into the 32 block secured area of the unit physical area P0 from the beginning thereof.

Next, if the user program 9 issues an input/output request requesting new writing of 7 block size data on the unit logical area L1 in FIG. The data is passed to the location processing means 6, and the reallocation processing means 6 executes processes S1 to S5 in FIG. 2 because new data is to be written.

Then, in process S2, since the current data size is 7 blocks, the reserved area size is determined to be 8 blocks, and in the next process S3, 8 blocks (0 ~7) is secured as a physical area of 8 block size to be used for writing the current data. At this time, the reallocation processing means 6 performs logical and
In the physical mapping management table 4, the unit logical area L1
number, unit physical area P1 number, and block position (
0 to 7), and also record the correspondence relationship with the unit physical area P.
Since the first 8 blocks of 1 have been secured, the contents of the free area management table 3 are updated accordingly.

After completing the above processing, the reallocation processing means 6 returns the physical area information including the number and block position (0 to 7) of the unit physical area P1 secured in step S3 to the input/output control means 2. (S5), input/output control means 2
is a unit physical area P for the magnetic disk device control means 7.
Issues an input/output request to block 1 (0 to 7). In response, the magnetic disk device control means 7 transfers the 7 block size data requested for new writing from the user program 9 to the unit physical area P1 of the magnetic disk device 8.
write to.

In this way, as shown in FIG. 2, the data of the 7 block size in the unit logical area L1 is written from the beginning to the reserved area of blocks 0 to 7 of the unit physical area P1.

Next, when the user program 9 issues an input/output request requesting new writing of data of one block size in the unit logical area L2 of FIG. 6, and the reallocation processing means 6 executes processes S1 to S5 in FIG. 2 because new data is to be written.

In step S2, since the current data size is one block, the secured area size is determined to be one block. Then, in the next process S3, the first block of the unit physical area P2, which is a free area, is secured as a physical area to be used for writing the current data. At this time, in the unit physical area P1, only one 8-block unit is secured, and the remaining 24 blocks are free areas, and it is also possible to secure one block from there, but in this embodiment, one physical Since areas of the same size are secured within the unit physical area of 32 block size corresponding to the area number, the first block of the unit physical area P2 is secured as described above.

[0030] Also, at this time, the reallocation processing means 6, in process S4, uses the secured unit physical area P.
2, record the correspondence between the number of the unit logical area L2, the number of the unit physical area P2, and the block position (0) in the logical-physical mapping management table 4, and further,
Since we have secured the first block of unit physical area P2,
The contents of the free area management table 3 are updated accordingly.

[0031] Upon completion of the above processing, the reallocation processing means 6 returns physical area information including the number of the unit physical area P2 determined in processing S3 and its block position (0) to the input/output control means 2. (S5), input/output control means 2
is a unit physical area P for the magnetic disk device control means 7.
Issue an input/output request to the 0th block of 2. In response to this, the magnetic disk device control means 7 writes the data of one block size requested for new writing from the user program 9 into the unit physical area P2 of the magnetic disk device 8.

In this way, as shown in FIG. 2, data of one block size in the unit logical area L2 is written to the reserved area of the 0th block of the unit physical area P2.

The method of using the physical area P up to this point is the same as the conventional method, but the following cases differ from the conventional method.

When the user program 9 issues an input/output request requesting new writing of 3 block size data in the unit logical area L3 in FIG. Since this is writing new data, the reallocation processing means 6 executes processes S1 to S5 in FIG.

In process S2, since the current data size is 3 blocks, the secured area size is determined to be 8 blocks. Then, in the next process S3, one 8-block unit is already secured, but the free area management table 3 manages the 8th to 31st blocks as free areas in the unit physical area P1. The 8 blocks from the 8th to the 15th are secured as one 8 block unit to be used for writing the current data. Note that at this time, the reallocation processing means 6
In process S4, based on the blocks of the unit physical area P1 secured above, the correspondence between the number of the unit logical area L3, the number of the unit physical area P1, and the block position (8 to 15) is added to the logical-physical mapping management table 4. Since the relationship has been recorded and eight blocks from the 8th to the 15th block of the unit physical area P1 have been secured, the contents of the free area management table 3 are updated accordingly.

When the above processing is completed, the reallocation processing means 6 sends the physical area information including the number of the unit physical area P1 secured in step S3 and its block position (8 to 15) to the input/output control means 2. The input/output control means 2 then issues an input/output request to the 8th to 15th blocks of the unit physical area P1 to the magnetic disk device control means 7 (S5). In response, the magnetic disk device control means 7 writes the 3-block size data requested for new writing by the user program 9 into the unit physical area P1 of the magnetic disk device 8.

In this way, as shown in FIG. 2, the 3-block size data of the logical area number L3 is stored in the reserved area from the 8th block to the 15th block of the unit physical area P1. More will be written.

Next, when the user program 9 issues an input/output request requesting new writing of data of one block size in the unit logical area L4 in FIG. 6, and new data is written, so processes S1 to S5 in FIG.
is executed.

Then, in process S2, since the current data size is 1 block, the secured area size is determined to be 1 block, and in the next process S3, one block unit has already been secured, but the first block The first block from the unit physical area P2, which is managed in the free area management table 3 as a free area thereafter, is secured as one block unit to be used for writing the current data. At this time, in process S4, the reallocation processing means 6 adds the number of the unit logical area L4, the number of the unit physical area P2, and its The correspondence with block position (1) is recorded, and the contents of the free area management table 3 are updated in accordance with the reservation.

[0040] When the above processing is completed, the reallocation processing means 6 returns the physical area information including the number of the unit physical area P2 secured in processing S3 and its block position (1) to the input/output control means 2. (S5), input/output control means 2
is a unit physical area P for the magnetic disk device control means 7.
2 issues an input/output request to the first block. In response to this, the magnetic disk device control means 7 writes the 1 block size data requested for new writing by the user program 9 to the unit physical area P2 of the magnetic disk device 8.

In this way, as shown in FIG. 2, data of one block size in the unit logical area L4 is written to the first block of the unit physical area P2.

After that, if the user program 9 issues an input/output request requesting new writing of 29 block size data in logical area number L5 in FIG. 2, the reallocation processing means 6 writes new data, so Processes S1 to S5 are executed, and in process S2, since the current data size is 29 blocks, the secured area size is determined to be 32 blocks. Then, the next process S3
Then, the 32 blocks of the unit physical area P3 are secured as a 32 block unit to be used for writing the current data, and in step S4, based on the unit physical area P3 determined above, the unit is set in the logical-physical mapping management table 4. The correspondence between the number of the logical area L5, the number of the unit physical area P3, and its block position (0 to 31) is recorded, and the contents of the free area management table 3 are updated according to the reservation.

[0043] When the above processing is completed, the reallocation processing means 6 returns the physical area information including the number and block position (0 to 31) of the unit physical area P3 secured in process S3 to the input/output control means 2. Then (S5), the input/output control means 2 issues an input/output request to blocks 0 to 31 of the unit physical area P3 to the magnetic disk device control means 7. In response, the magnetic disk device control means 7 transfers the 29 block size data requested for new writing from the user program 9 to the unit physical area of the magnetic disk device 8.
Write in 3.

In this way, as shown in FIG. 2, data of 29 block size in the unit logical area L5 is written from the beginning of the unit physical area P3.

Referring to FIG. 2, regarding data writing to a total of six unit logical areas from unit logical areas L0 to L5, in this embodiment, unit physical areas P0 to P3 are written.
A total of four unit physical areas are required. In contrast, in the conventional technology, a total of six unit physical areas are required, which deteriorates the usage efficiency accordingly.

Next, the operation when overwriting new data or adding new data to data already recorded on the magnetic disk device 8 will be explained.

For example, if the user program 9 issues an input/output request to overwrite or add new data of 4 block size to existing data of 3 block size in unit logical area L3 in FIG. The control is then passed to the reallocation processing means 6 along with the necessary information.

Since the reallocation processing means 6 is writing to existing data, it proceeds from processing S1 to processing S6 in FIG. The physical area is determined by converting the number of the area L3 into the number of the corresponding unit physical area P1 and its block position (8 to 15).

Next, the reallocation processing means 6 determines whether or not the existing reserved area size will be exceeded by data expansion (S7). In this case, the existing data with a size of 3 blocks is stored in units of 8 blocks, and the overwritten or additional data has a size of 4 blocks, so it can be stored in the original reserved area. Therefore, the reallocation processing means 6 determines in step S7 that the existing secured area size will not be exceeded, proceeds to step S5, and inputs and outputs the physical area information including the number of the unit physical area P1 and the block position (8 to 15). It is returned to the control means 2.

The input/output control means 2 issues an input/output request (overwriting or addition of data) to blocks 8 to 15 of the unit physical area P1 to the magnetic disk device control means 7. In response, the magnetic disk device control means 7 transfers the 4-block size data requested to be overwritten or added from the user program 9 to the unit physical area P1 of the magnetic disk device 8.
write to the physical area of

Although the size of the existing secured area was sufficient for writing new data into the existing data as described above, for example, the user program 9 writes, for example, a 10 block size to the 3 block size existing data in the unit logical area L3 in FIG. When an input/output request is issued to overwrite or add new data, the size of the existing reserved area is insufficient, and the following operations occur.

Since the reallocation processing means 6 is writing to existing data, it proceeds from processing S1 to processing S6 in FIG. After converting the data into the number and block position (8 to 15) of the unit physical area P1, it is determined whether or not the data will extend beyond the existing secured area (S7). In this case, the existing 3-block size data is stored in 8-block units, and the overwritten or additional data has a capacity of 10 blocks, so none of it can be stored in the original 8-block units.

[0053] Therefore, the reallocation processing means 6
A new secured area unit is determined according to the expanded data size (S8). In this case, the existing data has a size of 3 blocks and the overwritten or additional data has a size of 10 blocks, so the total size is 10 blocks or 13 blocks. Therefore, 32 blocks is determined as the secured area size.

After this determination, the reallocation processing means 6
refers to the free area management table 3 and secures a physical area of 32 block size for writing the expanded data (S9). For example, the usage status of the physical area is shown in Figure 2.
For example, if the unit physical area P4 is 3
Secure 2 blocks. Then, according to the secured unit physical area P4, the correspondence between the number of the unit logical area L3, the number of the unit physical area P4, and the block position (0 to 31) is recorded in the logical-physical mapping management table 4. , since 32 blocks of unit physical area P3 have been secured,
The contents of the free area management table 3 are updated accordingly (S10).

Next, the reallocation processing means 6 determines whether the current writing of existing data is overwriting or addition (S11). In the following, the case of overwriting and the case of addition will be explained separately.

(1) In the case of overwriting For example, as shown in FIG. 4, there is a write request for overwriting 10 block size data on existing 3 block size data on the magnetic disk device 8, and step S7 in FIG. If it is determined by the check that the size exceeds the existing secured area size, and if 32 blocks of the unit physical area P4 are secured as a new area in step S8, the reallocation processing means 6 proceeds to step S5 via step S13,
Physical area information including the number of the unit physical area P4 and the block position (0 to 31) is returned to the input/output control means 2. Upon receiving this return, the input/output control means 2 issues an input/output request (overwriting) to the unit physical area P4 to the magnetic disk device control means 7. In response, the magnetic disk device control means 7 writes the 10 block size data requested to be overwritten by the user program 9 into the unit physical area P4, as shown in FIG. At this time, the 3-block size area of the unit physical area P1 in which the data before being overwritten is recorded is no longer needed, so it is released by the reallocation processing means 6. This release is performed by deleting the correspondence between the unit logical area L3 number, unit physical area P1 number, and block position (8 to 15) from the logical-physical mapping management table 4, and free area management. This is done by returning the original secured area to a free area by updating table 3.

(2) In the case of addition For example, as shown in FIG. 5, there is a write request for adding 10 block sizes to the existing 3 block size data on the magnetic disk device, and the check in step S7 of FIG. When it is determined that the size of the reserved area exceeds the existing secured area size, and 32 blocks of the unit physical area P4 are secured as a new area in step S8, the reallocation processing means 6 first issues necessary instructions in step S12. The data is given to the transfer processing means 5 to perform the transfer process.

In response to this instruction, the transfer processing means 5 first transfers the unit physical area P1 through the magnetic disk device control means 7.
Read existing 3 block size data existing in the 8th to 10th blocks of Write to the second block. As a result, the original data to be added has been moved to the newly secured area as shown in FIG.

After that, the reallocation processing means 6
The process advances to process S5 via process S13, and physical area information including the number of unit physical area P4 and the block position (3 to 31) is returned to the input/output control means 2. Upon receiving this return, the input/output control means 2 issues an input/output request (overwriting) to the unit physical area P4 to the magnetic disk device control means 7. In response, the magnetic disk device control means 7 transfers the 10 block size data requested for additional writing from the user program 9 to the unit physical area P4, as shown in FIG.
Write from the third block onwards. Then, in step S13, the 3-block size area of the unit physical area P1 in which the data before additional writing was recorded is determined by the number of the unit logical area L3 and the unit physical area P1 from the logical-physical mapping management table 4. The correspondence between the number and the block position is deleted, and the free area management table 3 is updated to make the original secured area a free area.
To be released.

Next, the operation of reading data written in the physical area P of the magnetic disk device 8 as shown in FIG. 2 will be explained.

For example, if the user program 9 specifies the number of the unit logical area L3 and issues an input/output request for reading data as shown in FIG. The information is transmitted to the input/output control means 2, and control is passed to the reallocation processing means 6 along with necessary information.

Since the reallocation processing means 6 is reading data, it refers to the logical-physical mapping management table 4 and assigns the number of the unit logical area L3 passed from the input/output control means 2 to the corresponding unit physical area. It converts it into the number of P1 and the block position (8 to 15), and notifies the input/output control means 2 of it.

The input/output control means 2 provides the magnetic disk device control means 7 with the 8th to 15th areas of the unit physical area P1.
Issue an input/output request (read) to the block up to the block. In response to this, the magnetic disk device control means 7
Data in 8 to 15 blocks of the unit physical area P1 of the magnetic disk device 8 is read and stored in a buffer that can be referenced by the user program 9, for example.

[0064] By performing the above processing,
The user program 9 can refer to the data on the unit logical area L3 in the same manner as in the past.

[0065]

[Effects of the Invention] As explained above, the present invention sets multiple types of reserved area sizes set to the same size as the unit physical area and several smaller sizes, and selects a write request from among them. By determining the secured area size that is appropriate for the data size and securing the area on the magnetic disk device using that size, wasted space that was generated due to uniform area securing regardless of the data size is eliminated. Therefore, it is possible to improve the space efficiency on the magnetic disk device.

[Brief explanation of the drawing]

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

FIG. 2 is a diagram showing a configuration example of a logical area L of a file recognized by a user program 9 and a physical area P on a magnetic disk device 8, and a relationship between the two.

FIG. 3 is a flowchart showing an example of processing performed by reallocation processing means 6 when a write request is received from user program 9;

FIG. 4 is an explanatory diagram of operations when overwriting existing data.

FIG. 5 is an explanatory diagram of operations when additionally writing to existing data.

FIG. 6 is an explanatory diagram of operations when reading data.

[Explanation of symbols]

1...High-speed file management means 2...I/O control means 3...Free area management table 4...Logical/physical mapping management table 5...Transfer processing means 6...Reallocation processing means 7...Magnetic disk device control means 8...Magnetic disk device 9...User program

Claims (2)

[Claims] 1. A logical area of a file recognized by a user program is divided into unit logical areas of a plurality of block sizes, and a physical area on a magnetic disk device is divided into continuous unit physical areas of the same size as the unit logical area. ,
In a file system for an electronic computer that speeds up the input/output of large-capacity file data by expanding the unit of input/output into multiple blocks, a logical/physical interface that maintains the correspondence between the logical area and the physical area. A mapping management table, a free area management table for managing unused physical areas on the magnetic disk device, and a correspondence relationship between the logical area and the physical area using the logical-physical mapping management table. and managing the use state of the physical area using the free area management table, and upon a new write request for data specifying a logical area number from the user program,
Determine a secured area size suitable for the requested data size from among multiple types of secured area sizes set to the same size as the unit physical area and several smaller sizes, and determine the determined secured area size. A reallocation processing means that secures a physical area from an unused unit physical area or a partially used unit physical area with the same secured area size, and a physical area secured by the reallocation processing means that 1. A disk space management method comprising: input/output control means for writing data in a unit logical area corresponding to a given logical area number through a magnetic disk device control means. 2. When the user program requests overwriting or additional writing of data specifying a logical area number, the reallocation processing means changes the data size after overwriting or additional writing to the data before being overwritten or additionally written. Check whether the size of the physical area that has already been reserved for A physical area of a size suitable for the data size is secured from an unused unit physical area or a partially used unit physical area with the same secured area size, and when additional writing is performed, the original data is transferred to the newly allocated physical area. The input/output control means is notified of the newly secured physical area by transfer, and the original secured area is released, and the input/output control means requests the overwriting or additional writing to the notified physical area. 2. The disk space management system according to claim 1, wherein the data is written through said magnetic disk device control means.