JPH0568788B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a magnetic disk file access method, and in particular, a method for reading a directory file stored on a magnetic disk and accessing an individual file with an input/output request based on this information. Related to magnetic disk file access method.

[Conventional technology]

Conventionally, this type of magnetic disk file access method searches for the target individual file from a file containing list information of all individual files called a directory file each time an input/output request is received, as shown in FIG. From this information, the position of the target record on the magnetic disk of the target individual file is acquired, read, and input/output processing such as writing is performed.

That is, input/output processing is performed in the order of requests, in which the processing for one request is completed and then the next request is received.

[Problem that the invention seeks to solve]

The conventional magnetic disk file access method described above is configured to read the directory file on the magnetic disk every time an input/output request is received and perform input/output processing on the target individual file, so the number of accesses to the magnetic disk is reduced. increases,
Furthermore, when accessing a target record located far from the directory file, there is a disadvantage that the processing time increases because the disk head has to travel a long distance.

SUMMARY OF THE INVENTION An object of the present invention is to provide a magnetic disk file access method that can minimize the number of accesses to files on a magnetic disk and the distance traveled by the disk head.

[Means for solving problems]

The magnetic disk file access method of the present invention reads a directory file from a magnetic disk and stores it in a directory file area of a storage unit, stores the position of the disk head after reading the directory file in a disk head position area of the storage unit, File management including a file directory information storage area, an input/output request information save area, a work area, and a file configuration information storage area at predetermined positions of the storage unit corresponding to each individual file registered in the directory file. A block is secured and predetermined file configuration information is stored in the file configuration information storage area, and file directory information of each individual file of the directory file is stored in the corresponding file directory information storage area for a predetermined period of time. receives an input/output request for each individual file, stores information on each input/output request in the corresponding input/output request information storage area, and stores information on each individual file included in the information on each input/output request. The physical block number of the target record is calculated from the information stored in each of the file management blocks and stored in each work area, and when the number of individual files for which the input/output request is requested is plural, the number of each of these individual files is calculated. The distance between the position of the target record on the magnetic disk and the position of the disk head is calculated from the information stored in the file management block and the disk head position area, respectively, and the individual file of the target record located at the shortest distance is selected. When the number of individual files of the target record located at the shortest distance or the number of individual files with the input/output request is 1, the input/output processing of this individual file is stored in the input/output request information save area of these individual files. The disk head position stored in the disk head position area is checked each time each input/output process is completed.
It has a configuration that updates the position of the disk head after the input/output processing is completed.

〔Example〕

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

FIG. 1 is a flowchart of a program used in one embodiment of the present invention.

When the magnetic disk access service is started, first, in step _S1 , a directory file is read from the magnetic disk and stored in the directory file area of the storage section.

Next, in step _S2 , the disk head position after reading the directory file is stored in the disk head position area of the storage section.

Next, in step _S3 , from the information of the already stored directory file, a file management block is stored in the storage unit for each individual file that is recognized and distinguished by a specific file identification number registered in this directory file. to be secured. As shown in FIG. 2, this file management block includes a work area 11, an input/output information save area 12, and a file directory information storage area 1.
3 and a file configuration information storage area 14, and predetermined file configuration information is stored in the file configuration information storage area 14.

Next, in step _S4 , file directory information corresponding to each individual file from the stored directory files is stored in the file directory information storage area 13 in the file management block 1 of each individual file. This information will be maintained until the end of the service.

Next, in step _S5 , the input/output request for each individual file is received by the input/output request receiving queue for a predetermined period of time, and when the input/output request is received, the information of the input/output request is processed in step _S6 . The input/output request information is stored in the input/output request information save area 12 in the file management block 1 of the individual file (referred to as individual file A).

Next, in step _S7 , it is checked whether or not there are input/output requests for other individual files in the input/output request receiving queue, and if there are no input/output requests, the process proceeds to step _S8 .
However, if it exists, return to step _S6 . Here, let us consider the case where there is also an input/output request for individual file B. In this case, similar to the case of individual file A, input/output request information is transferred to individual file B.
The input/output request information is stored in the input/output request information save area 12 in the file management block 1.

After the input/output request has been processed for a predetermined period, in step _S8 , the physical block number of the target record of the individual file A for which the input/output is requested is stored as the input/output request information in the file management block 1 and the file directory information. , and the file configuration information, and work area 1 of individual file A.
Store in 1. Similarly, the physical block number of the target record of individual file B is calculated, and work area 1 in file management block 1 of individual file B is calculated.
Store in 1.

Next, in step _S9 , it is checked whether the number of individual files with input/output requests is multiple or one. If there are multiple files as in the above example, the process advances to step _S10 , and
When this happens, proceed to step _S11 .

In step _S10 , the distance between the position of each target record on the magnetic disk and the position of the magnetic disk is calculated based on the information stored in each file management block 1 and disk head position area. , determine the individual file of the target record that is the shortest distance away. For example, in the case of the individual files A and _B in the above example, as _shown in FIG _. , L ₂ is calculated, and in this case, since L ₂ is shorter, the individual file B is selected.

Next, in step _S11 , input/output processing is performed for the individual file determined in step ₁₀ , or for the individual file when there is only one input/output request, based on the information in the input/output request information save area 12 of each individual file. It is done.

After this input/output processing is completed, in step _S12 , the disk head position stored in the disk head position area is updated to the disk head position at this point in time.

Then, in step _S13 , it is checked whether all the input/output processing for the individual files requested for input/output has been completed during this predetermined period. If not, the process returns to step _S9 ; if it has been completed, the process returns to step S9.
In _S14 , it is determined whether or not to receive an input/output request for the next predetermined period. If it is received, the process returns to step _S5 , and if not, the job ends.

〔Effect of the invention〕

As explained above, the present invention saves the file directory information read at the start of the magnetic disk access service, stores various information about each individual file in the file management block, and stores the current location of each disk head individually. By updating the file each time file input/output processing is performed, and accessing the shortest target record based on this information, the number of accesses to the file on the magnetic disk and the distance traveled by the disk head can be minimized. This has the effect of reducing processing time and improving access efficiency.

[Brief explanation of drawings]

FIG. 1 is a flow chart of a program used in an embodiment of the present invention, FIG. 2 is a storage area configuration diagram showing an example of the storage area configuration of a file management block used in the present invention, and FIG. FIG. 4 is a schematic diagram for explaining the processing contents of step _S10 shown in FIG. 1, and FIG. 4 is a flowchart showing an example of a program used in the conventional magnetic disk access method. 1...File management block, 2...Magnetic disk, 3...Disk head, 11...Work area, 12...I/O request information save area, 13
...File directory information storage area, 14
...File configuration information storage area, 21 _a , 21 _b
...Objective record, _S1 to _S14 , _S21 to _S25 ...Step.

Claims (1)

[Claims] 1 Read the directory file from the magnetic disk and store it in the directory file area of the storage unit, store the disk head position after reading the directory file in the disk head position area of the storage unit, and store each directory file registered in the directory file. A file management block including a file directory information storage area, an input/output request information save area, a work area, and a file configuration information storage area is secured at a predetermined position of the storage unit corresponding to each individual file, and the file configuration information is stored in the storage unit. Predetermined file configuration information is stored in a storage area, file directory information of each individual file of the directory file is stored in the corresponding file directory information storage area, and input to each individual file is stored at a predetermined period. Receives an output request, stores the information of each input/output request in the corresponding input/output request information storage area, and sets the physical block number of the target record of each individual file included in the information of each input/output request. It is calculated from the information stored in each of the file management blocks and stored in each work area, and when the number of individual files for which the input/output request is made is plural, the purpose record of each of these individual files is stored on the magnetic disk. The distance between the position and the disk head position is calculated from the information stored in the file management block and the disk head position area, respectively, to determine the individual file of the target record located at the shortest distance, and the individual file located at the shortest distance is determined. When the number of individual files in the target record or the number of individual files with the input/output request is 1, the input/output processing of this individual file is performed using the information stored in the input/output request information save area of these individual files, and A magnetic disk file access method characterized in that the position of the disk head stored in the head position area is updated to the position of the disk head after each input/output process is completed each time the input/output process is completed.