JPH03225412A - File data read system - Google Patents

Abstract

PURPOSE:To improve the read processing speed by retrieving storage locations with a filing system consisting of plural storage devices without using the volume indexes of the storage devices. CONSTITUTION:A data position setting means 23 retrieves a filing system detail information control table 21 and a 1st retrieval key at this time is a partition ID and retrieval is performed with a next file name. The presence of file data on a data storage position control file 22 indicating the leading storage position is recorded as an address pointer in the table 21, so a target storage position record is accessed by using the pointer to extract from which block number in which storage device the file data of a file 'F' are stored. Then a data read means 24 reads the target file data out of storage positions of a specified storage devices 11-13 based on the information from the data position setting means 23. Consequently, the processing speed is improved on the whole.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a file data reading method, and more particularly to a file data reading method in a computer file system that uses a plurality of storage devices as auxiliary storage resources.

[Conventional technology]

Conventionally, when reading file data from file to file in a file system composed of multiple storage devices, first, the storage information of the file data provided in each storage device and stored in its own device is recorded. The volume catalog was searched, the storage location of the target file data was determined from the search results, and the file data was read from that location. That is, in principle, search processing is performed on a plurality of storage devices.

[Problem to be solved by the invention]

In the conventional file data reading method described above,
In order to identify the location of the target file data, in principle, it is possible to locate multiple files, not only when a file belonging to a file system is allocated to multiple storage devices, but also when it is contained in only one storage device. Since the read process for searching is repeated for the storage device,
This has the disadvantage that the overall reading processing speed decreases.

An object of the present invention is to solve the above-mentioned drawbacks and provide a file data reading method with high reading speed.

[Means to solve the problem]

The file data reading method of the present invention is a file data reading method for a file system composed of a plurality of storage devices. A data storage location management file in which storage location records expressed by number and block number are collectively recorded for each file in the storage order; (b) the file name of each file, the so-called alias of the file system to which the file belongs, and its a file system detailed management table (c
) data position setting means for extracting the storage position of each file data from the file system detailed management table and the data storage position management file; (d) reading actual file data based on information from the data position setting means; It has data reading means.

〔Example〕

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

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

The system shown in FIG. 1 includes three types of storage devices 11.1.
2.13, and a file system control section 2 that controls these storage devices, and processes input/output requests from the user program 3. The file system control unit 2 includes a file system detailed information management table 21 in which file management information of each file is recorded;
A data location where the storage location of the target file data is extracted from the data storage location management file 22 that records the storage location of actual file data in blocks, the file system detailed information management table 21, and the data storage location management file 22. an input/output control section that includes a setting means 23 and a data reading means 24 that reads actual file data from the storage location of the storage device instructed by the data position setting means 23, and further separates input/output requests from the user program 3; 25, and a data write control section 26 that controls and executes data writing.

FIG. 2 is a format diagram showing an example of the configuration of the file system detailed information management table 21, which is a virtual partition control method in which one file is optimally allocated on multiple storage devices to match the processing speed required by the user program. An example is shown. For the partition ID and file system ID, which are identification names for identifying an area called a partition set across multiple storage devices and the file system configured there, the file name of each file belonging to this file system. and individual file management information including an address file pointer indicating the address on the data storage location management file 22 where the storage location record indicating the first storage location of the file data is stored.

FIG. 3 is a format diagram showing an example of the structure of a storage location record of the data storage location management file 22. As shown in FIG. The storage location record is a fixed-length record that indicates the storage location on the storage device in units of plots as shown in the figure. They are arranged in the order in which they are stored, and the file is essentially a variable length record.

Next, the operation when the user program 3 reads, for example, X bytes from the beginning of a file whose partition ID is rVPJ and whose file name is "F" will be described. Commands from the user program 3 are divided into input processing and output processing by the input/output control section 25, and control of the read command is transferred to the data position setting means 23.

The data location setting means 23 first searches the file system detailed information management table 21. The first search key in this case is the partition ID (or file system ID)
), and then a search is performed by file name. In the file data detailed information management table 21, as shown in FIG. 2, the location of the storage location record indicating the storage location of the beginning of file data in the data storage location management file 22 is recorded as an address file pointer. Use this to access the desired storage location record,
Extract from which block number of which storage device the file data of file "F" is stored.

Next, the data reading means 24 reads the data position setting means 2.
Based on the information from step 3, the target file data is read from the specified storage location of the storage device.

If the number of bytes requested by the user program Since the data has been recorded, it is only necessary to read the storage position records consecutively for the number of blocks that satisfy the number of bytes X.

In addition, for requests to read from the middle of a file, by specifying the reading start position in bytes from the beginning of the file and the read size in bytes, the length exceeds the one block length mentioned above. File data can be read using processing similar to reading file data.

As mentioned above, the file system detailed information management table 21 and the data storage location management file 22 are used to determine the storage location, and the storage device is accessed only when reading data, so the overall processing speed is reduced. improves.

〔Effect of the invention〕

As explained in detail above, the present invention is a file system composed of a plurality of storage devices, and since the storage location is searched without using the volume catalog of the storage device, reading from the storage device is not necessary. Only the file data required is required, which has the effect of improving the reading processing speed.

[Brief explanation of drawings]

Figure 1 is a block diagram showing the system configuration of an embodiment of the present invention, Figure 2 is a format diagram of a file system detailed information management table, and Figure 3 is a format diagram of a storage location record in a data storage location management file. be. ■...Storage unit, 2...-File system control unit, 3...-User program, 11,
12゜13...Storage device, 21...File data detailed information management table, 22...
Data storage position management file, 23... Data position setting means, 24... Data reading means, 2
5...--input/output control section, 26...-data write control section.

Claims (1)

[Claims] In a file data reading method for a file system configured with a plurality of storage devices, (a) the storage position of file data in blocks is determined by the device number of the storage device, the cylinder number within the storage device, and the block; A data storage location management file in which storage location records expressed as numbers are recorded for each file in the order of storage; (b) the file name of each file, the identification name of the file system to which that file belongs, and the file of that file; (c) a file system detailed management table recording file management information including a pointer indicating a position on the data storage position management file of the storage position record corresponding to the first storage position of data; and (d) data reading means for reading actual file data based on information from the data position setting means. A file data reading method featuring: