JPH0442335A - Disk control system - Google Patents

Abstract

PURPOSE:To enable a user to control a large capacity file covering plural disk volumes with high efficiency by performing a file access after regarding a control file as a real file and retransferring automatically these files due to the overvolume. CONSTITUTION:A user regards a control file A' stored in a control file 12Aa as a real file A and has an access to a disk. The file A' stores the No (n) of a disk volume stored in the file A and therefore an OS 11a included in a CPU 11 has an access to a real file included in a volume (n) to return the writing result of the real data and the read data to the user as the results. Meanwhile the user can use an access interface to a normal file with no consciousness of the presence of a control file. Thus the user can control a large capacity file covering plural disk volumes with high efficiency.

Description

[Detailed Description of the Invention] [Summary] This invention relates to a disk management system that manages multiple disk volumes. In a system connected to volumes, one of the disk volumes is the management volume, and the other is the real volume.The real volume stores the actual data files, and the management volume stores the real files and one disk volume.
A management file that corresponds to one-to-one is provided, and this management file stores information such as whether the real file is on this volume or not, information of the user who is currently using the real file, etc. Configure the disk so that it can be accessed by treating it as a real file.

[Field of Industrial Application] The present invention relates to a disk management system that manages a plurality of disk volumes.

[Prior Art 1] FIG. 5 is a block diagram showing an example of the configuration of a conventional disk management system. A CPU I and a plurality of disk volumes 2 are connected via a bus 3. When accessing a large file spread across multiple volumes in such a system, the user must: (1) consciously determine in which disk volume 2 the file is located;

■When a certain disk volume 2 becomes full, the data is stored in another disk volume, but the user does this consciously.

It looks like this. In order to realize this kind of processing, ① Use an access interface for normal files.

It looks like this.

[Problem to be solved by the invention] For this reason, the user's program needs to skip the above ■ and ■ every time it accesses a file.
This results in a very complicated program. For example, even if a large-capacity file is spread across multiple volumes, if the file remains permanently in use, that is, if the file is never newly created or deleted, then ■ and ■ above will not be a problem. , when using the actual file,
Operations such as new file creation and deletion are constantly performed.

The present invention has been made in view of these problems, and it is possible to efficiently manage large-capacity files spanning multiple disk volumes when large-capacity files are arbitrarily created and deleted. We are aiming to create a disk management system that can

[Means for Solving the Problems] FIG. 1 is a block diagram of the principle of the present invention. Components that are the same as those in FIG. 5 are indicated by the same reference numerals. In the figure,
1-1 is a CPU, 12 is a CPU 1] and a plurality of disk volumes connected via a bus 3. CPUII
Inside, 0311a is provided for performing disk management control. The disk volume 12 is divided into two volumes, one of which is a management volume]2A, and the other part of which is a real volume 12B. 12Aa is a management file that corresponds one-to-one with the actual file. The management file 1.2Aa stores information on whether or not the real file exists in this volume, information on the user who is currently using the real file, and the like.

[Operation] FIG. 2 is an explanatory diagram of the operation of the present invention. The user accesses the disk by regarding the management file A' in the management file 12Aa as the real file A (■). Management file A
- stores the number n of the disk volume in which the real file A is stored, so the
S 11 , a attempts to access the real file in volume n (■), and returns the actual data write result and read data to the user as a result (■). During this time, the user is not aware of the existence of the management file and can use the normal file access interface. Therefore,
According to the present invention, it is possible to provide a disk management system that can efficiently manage large-capacity files spanning multiple disk volumes when large-capacity files are arbitrarily created or deleted. In addition,
Such control is performed by an interface function in the OS 1.1a of the CPU.

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

In the embodiment of the present invention, the management file], 2 A a is subjected to exclusive control so that multiple users can access the same real file at the same time (multiple □PEN). Also, when a disk volume becomes full while trying to create a real file, it searches for another free disk volume and creates the file there. In order to realize such a function, the management file 12Aa is
The configuration is as shown in the figure.

This management file 12Aa has a one-to-one correspondence with real files existing in the real volume. When using a file exclusively, only one file descriptor is required for the actual file, but multiple open (without exclusive control)
The configuration is such that it can store up to 64 file disks in consideration of the case where the file is opened (OPEN). Furthermore, even if the volume capacity is exceeded when writing to a real volume, if there is space in another volume, the real file is moved there and processing continues. Therefore, since there is a possibility that the file disk libter of a real file may change even after opening, it is necessary to store the value at the time of opening and the current value.

In the figure, area 1 stores the current real file storage destination volume number. Area 2 is a multiple open bit assignment and stores 1 to 64 multiple open flags. When this flag is “O”, it is not open and “1”.
'', it is being opened. Area 3 is the file disc libter of the management file, which is stored at the time of opening and is fixed. Area 4 is the file disc libitor of the current real file, and the value of the current real file is Stored.

Area 5 contains the flag 1 reserve and actual file storage destination volumes. When the flag is "0", the storage location has not changed; when it is "1", it indicates that the storage location of the real file has been changed by another process. The storage destination volume number is entered. Areas 3 to 5 show records for one open. Areas 6 to 8 are records for 64 1'' processes.

FIG. 4 is a diagram showing an example of a sequence according to the present invention. In the figure, the issuance order of file interfaces when two users (programs), User 2 and User 2, have exclusive access to a real file, the contents of the management file at that time, and the LII other system system. - It shows the change in table capacity. Here, 1.2 A b is an exclusive control table (e n q) that manages who is using which table, and is composed of a file ID and a counter.

When user 1 opens the file, fdl is added to fd in the user's management file (■). Then, this number fdl is notified to user 1. If there is a request to lock and use the file, the user 1 sets N to the file ID in the exclusive control table 12Ab (■). This indicates the file number or N to which you want to lock the file. At the same time, set the counter to 1.

Then, writing to file N is performed (■).

If volume 3 becomes full during a write, the interface function moves the actual file to volume 4 and continues writing. On the other hand, user 2 also opens the file (■) and locks the file (■). This results in
The counter of the exclusive control table 12Ab is set to 2. However, since file N has been locked by user 1, it is waiting for the file to be unlocked (■).

When user 1 finishes writing to file N, the file lock is released (■) and the file is closed (■).
. User 2 waits for user 1 to release the lock and writes to file N (■).

In this case, even if an attempt is made to write, since the flag in the management file 12A, a is 1, it is determined that the currently open real file is old, the real file in the new volume 4 is re-opened, and the write process is performed. When the process is finished, release the lock [phase]) and close the file (■). In sequence (3), when the lock is released, the counter changes from 2 to 1, and when the lock is further released in sequence [phase], the counter becomes 0. When the counter becomes 0, the file ID also becomes 0.

[Effects of the Invention] As described above in detail, according to the present invention, a disk volume is divided into a management volume that stores management files and a real volume that stores data, and disk management is performed using an interface function. By doing so, ■The user accesses the managed file by regarding it as a real file.

■Transferring due to volume overload is done automatically.

■The file access interface is no different from a normal file access interface.

This makes it possible to efficiently process large files that span multiple disk volumes.

[Brief explanation of the drawing]

Fig. 1 is a principle block diagram of the present invention, Fig. 2 is an explanatory diagram of the operation of the present invention, Fig. 3 is a diagram showing a configuration example of a management file, Fig. 4 is a diagram showing an example of a sequence according to the present invention, Fig. 5 The figure is a block diagram showing an example of the configuration of a conventional disk management system. In Figure 1, 3 is a bus, and 1] is a CPU. 11a is 081, 1.2A is a management volume, and 12B is a real volume.

Claims (3)

[Claims] (1) CPU (11) and multiple disk volumes (1)
2), one of the disk volumes (12) is the management volume (12A), and the other is the real volume (12B).
), the actual data file is stored in the real volume (12B), the management file (12Aa) is provided in the management volume (12A) in one-to-one correspondence with the real file, and the management file (12Aa) is stores information on which volume the real file is located on, information on the user currently using the real file, etc., and allows the user to access the disk by regarding the management file (12Aa) as a real file. A disk management system configured as follows. (2) The CPU (1
2. The disk management system according to claim 1, wherein the disk management system is provided as a file access interface (function) in 1). (3) The disk management system according to claim 1, wherein the management file (12Aa) is managed under exclusive control so that a plurality of users can access the same real file at the same time.