JPH04205518A - Disk device control system - Google Patents

Abstract

PURPOSE:To carry out the read processing of a disk device at a high speed by issuing the same read requests to the multiplexed disk devices that receive the data of the same contents from a host system and adopting the first answer as the read result. CONSTITUTION:The data of the same contents received from a host system 3 are stored in the disk devices 11-1n respectively. When these stored data are read into the system 3, the same read request commands are issued to the disk controllers 21-2n from the system 3. Then the devices 11-1n return the data to the system 3 in response to the read request commands. The first returned data is adopted as the result of the read request command. Thus it is possible to prevent the dependence of the read time on the access time of a specific disk device and to extremely shorten the processing time.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a method for controlling a disk device, and more specifically, a method for controlling a disk device, which includes a plurality of disk devices and a plurality of disk control devices, and multiplexes information paths. This invention relates to a control method for a disk device.

[Prior art] A disk control device is a central processing unit (host system).
is located between the channel and the disk device, and controls the disk device based on instructions from a predetermined program. Also,
Since the disk control device serves as a channel and a path for common information between a plurality of disk devices, a failure of the device can make the entire subsystem inoperable, and may even cause the entire system to go down. Therefore, in systems that require particularly high reliability, information paths are multiplexed using a plurality of disk control devices in order to achieve high reliability.

As a conventional control method for this type of disk device, there is a system configuration as shown in FIG. 4, for example.

In this figure, reference numeral 11 denotes a host system consisting of a central processing unit (CPU), etc., which controls the entire device and performs predetermined arithmetic processing.
4 and exchanges commands and data with the disk controllers 13 and 14. The disk controllers 13 and 14 receive commands and address data from the host system 11 and control the plurality of disk devices 1.
Controls the recording and playback operations of 5.16. Disk device 1
Reference numeral 5.16 denotes a peripheral storage device consisting of, for example, a floppy disk device, a fixed disk device, an optical disk device, etc. The disk devices 15 and 16 are controlled by the disk controller 13.14 and store data sent from the host system 11. , and the disk device 15
．． The recorded data recorded in the host system 16 is transferred in blocks to a main memory device (for example, a semiconductor memory element) in the host system 11 and processed by the central processing unit. The above disk controller I4 and disk device 1
6 constitutes a backup disk system 17 as a whole, and the backup disk system 17 performs backup to supplement the failure of the disk device 15 and disk controller 14. in this case,
In order to ensure that the disk devices 15 and 16 have the same data, only updating (writing) is performed on each multiplexed disk device 15, 16.

[Problem to be Solved by the Invention] However, in such a conventional disk multiplexing control method, there is a mode in which only updating is performed so that each multiplexed disk device has the same data. Therefore, although it is possible to back up disk devices, the current situation is that functions other than the backup function, such as shortening access time, are not considered. In other words, in the conventional disk multiplexing control system, reading is performed only from a specific disk device (disk device 15 in FIG. 4), and reading is performed only from a specific disk device (the disk device 15 in FIG.
In FIG. 4, the disk device 16) is a specific disk device 1.
Because the system only functioned as a backup system that functioned only when the disk drive 5 failed, the read read time was limited to the specific disk drive 1.
The problem is that it depends on the access time (rotation wait, seek time) of 5.

This is due to the fact that disk devices and disk control devices are multiplexed just to realize the backup function, and in particular, reading is performed only from a specific disk device.

It is clear that if reading is performed for each multiplexed disk device, there will be no dependence on the access time of a particular disk device.

An object of the present invention is to enable reading from each multiplexed disk device.

[Means for solving problems] The means of the present invention are as follows.

Disk device 1. ~In (n represents a plurality) (see the block diagram in Figure 1, the same applies hereinafter) is a disk device such as a floppy disk device or a fixed disk device, and is of the same type with the same characteristics such as specifications and capacity. A plurality of disk devices are preferred.

Disk control device 2. ~2n is a plurality of disk devices 1
．． A plurality of disk controllers for respectively managing and controlling ~in.

The host system 3 controls the entire device and also controls a plurality of disk control devices 2. -2n, and sends commands and data to disk controller 2. It stores data in the disk drives 1 and -In through the disk drives 15 to 2n, or reads data from the disk drives 15 to 1n, and is composed of a central processing unit (CPU) and the like. In this case, the host system 3 causes the disk devices to store data of the same content, and when reading data stored in the disk devices 11 to In, the host system 3 stores data of the same content in the disk devices, and when reading data stored in the disk devices 11 to In, the host system 3 uses the plurality of disk control devices 2. The same read request command is issued to each of .

[Effect] The operation of the means of the invention is as follows.

First, disk device 1. ~1n stores data with the same content sent from the host system.

When reading data stored in the disk devices 11 to In into the host system 3, the host system 3 reads the data stored in the plurality of disk control devices 2. The same read request command is issued to each of the disk drives 1. When data is returned from ~1n to the host system 3, the first responded data among the returned data is adopted as the result of the read request command.

In other words, when reading data, multiple disk devices 1. The data from the disk device with the shortest access time among ~1n will be adopted,
The read time is prevented from being dependent on the access time (rotation wait time, seek time) of a specific disk device, and the processing time is significantly reduced.

Therefore, it becomes possible to read data from each multiplexed disk device.

[Example code] Examples will be described below with reference to FIGS. 2 and 3.

FIGS. 2 and 3 are diagrams showing one embodiment of a control method for a disk device.

First, the configuration will be explained. FIG. 2 is an overall configuration diagram of the disk device multiplexing control system. In this figure, 21
2 is a host system that performs various processes including read processing shown in FIG. 3, and the host system 21 is composed of a central processing unit (CPU) and the like. A plurality of disk controllers (disk control devices) 23 and 24 that manage multiplexed disk devices are connected to the host system 21 via a control bus 22. The same read requests are issued to the disk controllers 23 and 24 (read requests ■-1, ..., ■-
(see n). This read request is made asynchronously, and a response is returned from the disk controllers 23 and 24 (response -1, . . . ).

The host system 21 can execute other processes up to (see (2)-n). The disk controllers 23 and 24 perform read processing on each disk device 25 and 26 based on read requests ■-1 to ■-n from the host system 21 (read requests ■-1, ..., ■). -n), upon receiving a reading completion response from the disk device 25.26 (see ■-1,..., ■-n), returns a reading response to the host system 21 (response ■-1).
,..., see ■-n). Each disk device 25.26
As in the conventional example, it is composed of, for example, a floppy disk device, a fixed disk device, an optical disk device, etc., and each disk device 25, 26 is connected to a disk controller 23.
24, the data sent from the host system 21 is updated so that it has the same content. In response to the same read request issued from the host system 21 to each disk device 25, 26, the disk devices 25 and 26 having the same data are sent to the host system through the disk controller 23 and 24. Transfer to 21.

Next, the operation of this embodiment will be explained.

FIG. 3 is a flowchart showing data read processing in the host system 21 and disk controllers 23, 24, and the reference numeral 5n (n=1.2...
) indicates each step of the flow.

When the read process of reading stored data from the disk devices 25 and 26 starts in the host system 21,
First, in step S1, a count of disk multiplicity corresponding to the number of disk controllers and disk devices is set, and in step S2, the count is determined. When the count is greater than O (count > O), it is determined that there is a target disk controller to which a read request should be made, and a read request is output to the target disk controller in step S3, and in step S4. The count is decremented (count -1) and the process returns to step S2. If the count is not greater than O in step S2 (that is, the count is 20), it is determined that there is no target disk controller to which a read request should be made, and a read request is made first as shown in FIG. 3(B). Execute another process (however, only when another process exists) until a response is returned from the disk controller 23.24, and if there is a response from the disk controller 23.24, request to read the first response. This is adopted as the read result for , and the processing of this flow ends.

On the other hand, when a read request is sent to the target disk controller 23 in step S3, the program shown in FIG. Analyzes the command contents. Next, in step S12, since the analysis result in this case is a read request, a read request to the disk device 25 (Fig. 2 -
1)), and in step S13, the result of the read request (see Figure 2 -1) is transferred from the disk controller 23 to the host system 21 based on the response (see Figure 2 -1), and the disk controller 23 Finish the processing in.

Further, in step S3, when a read request is made to the disk controller 24, the program shown in FIG. 3(D) is started, and in step S21, the content of the command sent from the host system 21 is analyzed. Next, in step S22, since the analysis result in this case is a read request, a read request is made to the disk device 26 (see Fig. 2 -n), and in step S23, the response (Fig. 2 -
The result of the read request (see (2)-n in FIG. 2) is transferred from the disk controller 24 to the host system 21 based on the result (see (2)-n in FIG. 2), and the processing in the disk controller 24 is completed.

By executing the above processing program, the disk device multiplexing control system shown in FIG. 2 is realized. Here, the host system 21 has each disk controller 23 .
By uniformly issuing update commands to 24, the update (write) process is made to have the same data, and this update process itself is the same as in the conventional example.

[Effects of the Invention] According to the present invention, the host system issues the same read request to multiplexed disk devices having the same data, and the first response is adopted as the read result. As a result, it is possible to prevent dependence on the current state such as the specs of a specific disk device or the position of the head, and it is possible to speed up the reading process.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of the present invention, Figs. 2 and 3 are diagrams showing an embodiment of a control method for a disk device, Fig. 2 is a block diagram thereof, and Fig. 3 is a diagram of its read processing. Flowchart showing the program. FIG. 4 is a block diagram showing a conventional disk device control method. 11...Host system, 12...Control bus, 13.14...Disk controller (disk control device), 15.16...Disk device. Patent applicant Casio Computer Co., Ltd. Figure 1 2n In Figure 4■

Claims (1)

[Scope of Claims] A computer system comprising: a plurality of disk devices that store predetermined data; a plurality of disk control devices that control the disk devices; and a host system that exchanges commands and data with the disk control device; The plurality of disk devices store data of the same content sent from the host system, and when reading the data stored in the disk device into the host system, the plurality of disk control devices are sent from the host system to the plurality of disk control devices. When data is returned from the disk device to the host system in response to the read request command, the data that was responded to first among the returned data is used as the read request command. A control method for a disk device characterized in that it is adopted as a result of.