JPH07201132A - Duplex disk device and head position changing method - Google Patents

Abstract

PURPOSE:To improve the utilization efficiency and reliability of a central processing unit(CPU) by copying the storage data of a disk device to another disk device through the respective communication means of first and second disk controller connected to the CPU. CONSTITUTION:The CPU 1 sends the write or read request of data through a system bus 3 to the disk controller 2. The communication means 2a receives the read requests of the data and sends this command to a control means 2c, the means 2c reads out the data of the disk device 4 and sends them to a temporary storage means 2b and the communication means 2a reads out the data of the means 2b and sends them through the bus 3 to the CPU 1. Then, the controller 2 which receives a write command and the data from the CPU 1 writes the data in the device 4, sends the copy of the command and the data through the communication means 2b and a sub bus 7 to the disk controller 5 and writes the data in the disk device 6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dual disk having first and second disk control devices which are connected to a central processing unit and store the same data written from the central processing unit in respective controlling disk devices. It relates to the device.

[0002]

2. Description of the Related Art Conventionally, this type of dual disk device is
For example, it is used in a power supply control device that monitors and controls the power system. Even if one disk device fails, the data stored in the other disk device is used to continuously supply power, and a stable power supply to consumers is achieved. Power can be supplied.

FIG. 23 is a block diagram of a conventional duplex disk device called a duplex disk device.
Reference numeral 01 denotes a central processing unit, which is an operating system (hereinafter referred to as OS) 8 having a duplex driver 803.
It is equipped with 02. Reference numeral 804 is a first disk controller connected to the central processing unit 801 via the system bus 809, 806 is a disk apparatus in which data is written and read by the first disk controller 804,
Reference numeral 805 denotes the central processing unit 80 via the system bus 809.
The second disk controller connected to the first disk unit 807 is the second disk controller.
The disk control device 805 is a disk device for writing and reading data.

The first disk control unit 804 has a communication means 804a for communicating with the central processing unit 801 and the central processing unit 8.
01 temporary storage means 804b for temporarily storing the instructions and data, and disk device control means 8 for controlling the writing and reading of data to and from the disk device 806.
04c. The second disk control device 805 has the same configuration, and the communication means 805
a, temporary storage means 805b, disk device control means 80
5c.

Next, the operation will be described. FIG. 24 is a flow chart for explaining the data write operation. First, after the start, a write command (command to write data to the disk device, command to send a control command to the disk controller) from the central processing unit 801 and data size (disk Write data consisting of the length of input / output data with the device and the length of a control command sent to the disk controller) and data (data to be written to the disk device or a control command to be given to the disk controller) The communication means 804a of the device 804 receives it (step ST24-1). The communication unit 804a sends an instruction to the disk device control unit 804c and sends data to the temporary storage unit 804b (step ST24-2). The temporary storage means 804b temporarily stores the data (step ST
24-3).

The disk controller 804c interprets the command and controls the disk device 806. Further, the disk control device 804c, based on the command, temporarily stores means 804.
The data is extracted from b and written in the disk device 806 (step ST24-4).

After the above processing is completed, the central processing unit 80
1 sends the same command as above to the second disk controller 805. The second disk control device 805 writes data in the disk device 807 by the same operation as the first disk control device 804 (step ST24-
5).

FIG. 25 is a flow chart for explaining the data read operation. First, the communication means 804a of the first disk controller 804 receives read data consisting of a read command from the central processing unit 801 and a data size (step). ST25-1). This communication means 804
The a sends a command to the first disk controller 804c (step ST25-2).

The first disk control device 804c interprets the command, reads the data from the disk device 806, and sends the data to the temporary storage means 804b (step ST
25-3). Then, the communication unit 804a sends the data read from the temporary storage unit 804b to the central processing unit 801 (step ST25-4).

The central processing unit 801 is a disk controller 8
04, when the disk device 806 has failed, a read command is sent to the second disk control device 805 to perform the same operation, and the data read from the temporary storage means 805b is sent to the central processing unit 801.

FIG. 26 is a block diagram of a conventional duplicated disk device called a mirror disk device. The same parts as those in FIG. 26 are designated by the same reference numerals and their duplicate description will be omitted. In FIG. 26, reference numeral 808 denotes a disk control device connected to the central processing unit 801 via a system bus 809, which is a communication unit 808a for communicating with the central processing unit 801 and temporarily stores instructions and data written from the central processing unit 801. Temporary storage means 808b, double writing control means 808
c, writing of data to the disk device 806 by an instruction received via the double writing control means 808c,
This is a configuration including disk control means 808d for controlling reading and disk control means 808e for controlling writing and reading of data with respect to the disk device 807.

Next, the operation will be described. FIG. 27 is a flow chart for explaining the data write operation. First, the communication unit 808a of the disk controller 808 receives the read command and data from the central processing unit 801 (step ST27-1). The communication means 808a sends an instruction to the double writing control means 808c and sends data to the temporary storage means 808b (step ST27-2). This temporary storage means 808b temporarily stores the data (step ST27-3).

The double writing control means 808c interprets the command, controls the disk device control means 808d, and sends the data read from the temporary storage means 808b to the disk device control means 808d (step ST27-4).
The disk device control means 808d controls the disk device 806 according to an instruction from the double writing control means 808c and writes the data received from the double writing control means to the disk device (step ST27-5).

Next, the double writing control means 808c interprets the command to control the disk device control means 808e, and based on this command, sends the data read from the temporary storage means 808b to the disk device control means 808e (step ST27-6). The disk device control means 808e controls the disk device 807 in accordance with a command from the double writing control means 808c, and writes the data received from the double writing control means to the disk device (step ST27).
-7).

FIG. 28 is a flow chart for explaining the data read operation, in which the communication means 808a receives an instruction from the central processing unit 801 (step ST28-1). This communication means 808a sends a command to the double writing control means 808c (step ST28-2). The double writing control means 808c interprets the command and the disk device control means 80
The command is sent to 8d (step ST28-3).

The disk device control means 808d reads data from the disk device 806 and sends this data to the double writing control means 808c (step ST28-4).
The double writing control means 808c sends the sent data to the temporary storage means 808b and temporarily stores it (step ST
28-5), the temporary storage means 8 via the communication means 808a.
The data read from 08b is sent to the central processing unit 801 (step ST28-6).

An example of a conventional duplex disk device called the above-mentioned mirror disk device is disclosed in, for example, Japanese Patent Laid-Open No.
165579, JP-A-4-241016,
There is one disclosed in Japanese Patent Laid-Open No. 4-256121.

[0018]

As described above, FIG.
According to the conventional dual disk device shown in FIG. 1, the central processing unit has to perform the same data writing operation twice to the first and second disk control devices, and the central processing unit is operated during this writing operation. Not available for other processing. That is, there is a problem that the utilization efficiency of the central processing unit is low.

On the other hand, according to the conventional dual disk device shown in FIG. 26, since the central processing unit does not have to perform the same data write operation twice, the utilization efficiency of the central processing unit is higher than that of the conventional device. However, there is a problem that when the disk control device fails, the entire redundant disk device cannot be used.

Further, in any of the above conventional devices,
After replacing the failed disk unit, a data recovery process for storing data in this new disk unit and double data writing for storing the same data in the disk unit controlled by the first and second disk unit control means. Processing is performed independently without being aware of each other.
For this reason, if normal data is written to the new disk device by an interrupt during data recovery processing, data recovery processing is also performed on areas that do not need to be recovered due to the normal data being written, and data recovery processing efficiency is improved. There was a problem that was low.

Further, in the conventional dual disk device, in order to speed up the data reading process, it is judged which disk device head is closer to the track position of the data to be read, and the head is located closer. I was reading data from a disk device. By this process,
The seek time to move the head to the track position of the data to be read can be shortened, and the data reading speed can be increased.

However, in a duplicated disk device consisting of two disk devices in which the same data is stored, the final head track position is the same, for example, when the data writing process is completed. Therefore, when speeding up the above-described data reading process, the track speeds of the two disk devices are the same, and therefore the reading speed does not change regardless of which disk device is read.

The present invention solves the above problems, and an object of the present invention is to improve the utilization efficiency of the central processing unit and improve the reliability.

It is an object of the present invention to improve the efficiency of double data writing processing.

An object of the present invention is to improve the utilization efficiency of the central processing unit.

It is an object of the invention of claim 4 that a plurality of central processing units share one disk device to enable centralized management of data.

It is an object of the present invention to improve the efficiency of data recovery processing for storing data in a new disk device after replacement.

It is an object of the present invention to improve the data read speed from the disk device.

It is an object of the present invention to easily and surely move the heads of the disk device to different track positions after writing the data.

[0030]

According to another aspect of the present invention, there is provided a duplicated disk device, in which the communication means between the disk control devices is provided to each of the first and second disk control devices and the communication means. A disk copy means for copying the data stored in one disk device to the other disk device is provided.

According to a second aspect of the present invention, there is provided a dual disk device having a simultaneous write function means for writing the data to the other disk device when the disk controller writes the data to the disk device managed by itself. is there.

According to a third aspect of the present invention, there is provided a dual disk device in which each of the first and second disk control devices is provided with temporary storage means for temporarily storing the data written from the central processing unit. .

According to a fourth aspect of the present invention, there is provided a dual disk device in which the first and second disk control devices are connected to a plurality of central processing units.

According to a fifth aspect of the present invention, there is provided a duplicated disk device in which a central processing unit is provided with processing area management means for managing the processing area of the disk device in the dual writing process and the data recovery process.

According to a sixth aspect of the present invention, there is provided a dual disk device as a processing area management means, which manages the entire disk device in logical block units, and sets the processing area of the disk device in the dual writing process and the data recovery process to a logical block. A disk management table for each unit is provided in the central processing unit.

According to a seventh aspect of the present invention, there is provided a dual disk device as a processing area managing means in which a register for holding the position of the disk device currently undergoing recovery processing is provided in the central processing unit during data recovery processing. .

According to an eighth aspect of the present invention, in a duplicated disk device, when the head track positions of the disk devices controlled by the first and second disk control devices are equal, the head position of one disk device is changed to the other. The head position changing means for changing the head position of the disk device is provided in the central processing unit.

According to a ninth aspect of the present invention, there is provided a duplicated disk device having a head position changing means for changing a head position of a disk device which has finished writing data to a position different from a head position of the other disk device. It is provided in the processing device.

According to a tenth aspect of the present invention, there is provided a duplicated disk device having a head position changing means for changing a head of a disk device which has finished writing data to a predetermined position different from a head position of the other disk device. It is provided in the processing device.

In the head position changing method according to the present invention, the head position of each disk device is read by comparing the track positions corresponding to the heads of the disk devices to which the data writing process has been performed, from the track information register. When the corresponding track positions are the same, the track position at which the head of one of the disk devices should be changed is calculated, and the head is moved to the calculated track position.

According to a twelfth aspect of the present invention, in the head position changing method, the track position corresponding to the head of the disk device for which the data writing process has been completed is read from the track information register, and the head is changed based on the track position. The track position to be changed is calculated, and the head is moved to the calculated track position.

The head position changing method according to the thirteenth aspect of the present invention is to move each head of the disk device to a predetermined track position after the data writing process is completed.

[0043]

According to the first aspect of the present invention, there is provided a disk copy means for copying data stored in one disk device to another disk device via the communication means between the disk control devices so that the disk control devices can communicate with each other. When performing the data transfer operation, the occupation of the central processing unit and the system bus can be eliminated, and the utilization efficiency and reliability of the central processing unit are improved.

In the disk controller according to the second aspect of the present invention, when the data is written in the disk device controlled by itself, the write function means simultaneously writes the data in the other disk device through the disk copy means and the communication means. As a result, the double data writing process can be speeded up.

In the third aspect of the present invention, the temporary storage means temporarily stores the data written from the central processing means, and the data is written to the disk device by reading it from the temporary storage means. Data transmission time to the control device can be shortened,
The utilization efficiency of the central processing unit can be further improved.

Since the first and second disk control devices according to the fourth aspect of the present invention are connected to a plurality of central processing units, each central processing unit shares the disk unit and enables centralized management of data. To do.

The processing area management means in the fifth aspect of the present invention records the processing area in which the double data writing processing or the data restoration processing has been performed, and the double writing of the data must be done with reference to the recorded contents. Data recovery processing, and if recovery processing is completed, double writing
The efficiency of data recovery processing can be improved.

According to the disk management table of the sixth aspect of the present invention, the processing area of the disk device in the double writing process and the data recovery process is held for each logical block unit, for example, a plurality of sectors for each track of a cylinder. By referring to the recorded contents of this disk management table, if the data double writing has not been completed, the data recovery processing is performed, and if the recovery processing has been completed, the double writing is performed to improve the efficiency of the data recovery processing. be able to.

The register according to the invention of claim 7 is
By holding the position of the disk device that is currently undergoing recovery processing, the contents stored in this register are referenced. If data recovery processing has been completed, double writing is performed. Can be written to improve the efficiency of data recovery processing.

According to the eighth aspect of the invention, when the head track positions are the same, the head position changing means changes the head position of one of the disk devices to the head position of the other disk device to read the data from the head. The seek time for moving to the position is shortened, and the data read speed can be increased.

According to the ninth aspect of the invention, the head position changing means changes the head position of the disk device in which the writing of data is finished to a position different from the head position of the other disk device. The processing can be performed efficiently.

According to the tenth aspect of the invention, the head position changing means changes the head of the disk device, which has finished writing the data, to a predetermined position so that it is not necessary to judge the changed position of the head. The head position changing process can be easily performed.

In the head position changing method according to the present invention, it is judged that data has been written to the disk device controlled by each of the first and second disk control devices, and each disk device is judged. When the corresponding track positions of the two heads are the same, it is possible to move the heads of the respective disk devices to different track positions by calculating the track positions to which the heads of one of the disk devices should be moved. It can be done easily and reliably.

According to the twelfth aspect of the present invention, in the head position changing method, it is judged that data has been written to the disk device controlled by each of the first and second disk control devices, and the data is first written. By calculating the track position to move the head of the disk device that has completed the writing process and moving the head to the track position, it is possible to quickly and surely move the head of each disk device to different track positions. It can be carried out.

According to a thirteenth aspect of the present invention, in the head position changing method, it is judged that data has been written to the disk device controlled by each of the first and second disk control devices, and the data is written. After the end, by moving each head of the disk device to a predetermined track position, it is extremely easy to move the head of each disk device to a different track position.

[0056]

【Example】

Example 1. FIG. 1 is a block diagram showing an embodiment of the invention described in claim 1. In FIG. 1, 1 is a central processing unit,
Reference numeral 2 is a first disk controller connected to the central processing unit 1 via a system bus 3, 4 is a disk device in which data is written and read by the first disk controller 2, and 5 is a system bus 3. A second disk controller 6 is connected to the central processing unit 1 via a central processing unit 1, and a disk device 6 is used for writing and reading data by the second disk controller 5.

The first disk control unit 2 communicates with the central processing unit 1a, a temporary storage unit 2b for temporarily storing instructions and data written from the central processing unit 1, and a data writing to the disk unit 4. , A disk device control means 2c for controlling reading, a communication means 2d between the disk control devices, and a disk copy means 2e for copying data stored in one disk device to the other disk device via the communication means 2d. is there.

The second disk control device 5 has the same structure and has a communication means 5a, a temporary storage means 5b, a disk device control means 5c, a communication means 5d between the disk control devices, and a disk copy means 5e. is doing. And
Communication means 2 in the first and second disk control devices 2 and 5
d and 5d are connected by a sub bus 7.

Next, the operation of the first embodiment will be described. When the disk device 4 is currently used and the disk device 6 is reserved, the central processing unit 1 sends a data write or read request to the disk controller 2 via the system bus 3.

The disk controller 2, which has received the data read request from the central processing unit 1, reads the data from the disk unit 4 by the operation shown in the flowchart of FIG. 2 and sends the read data via the system bus 3. And sends it to the central processing unit 1.

The data read operation will be specifically described below with reference to the flow chart of FIG. First, when the communication means 2a receives a data read request (command) from the central processing unit 1, this command is sent to the disk device control means 2c.
(Step ST2-1).

The disk device control means 2c interprets the command, reads the data from the disk device 4, and sends this data to the temporary storage means 2b (step ST2-2). The communication means 2a reads data from the temporary storage means 2b,
This data is sent to the central processing unit 1 via the system bus 3 (step ST2-3).

Next, the disk controller 2 which has received a data write request (command) and data from the central processing unit 1
Writes data to the disk device 4 by the operation shown in the flowchart of FIG.
A copy of the above command and data is sent to the disk controller 5 via the communication means 2d and the sub bus 7. The disk controller 5 writes data to the disk device 6.

The data write operation will be specifically described below with reference to the flowchart of FIG. First, when the communication means 2a receives a data write request (command) from the central processing unit 1 (step ST3-1), it sends this command to the disk device control means 2c and sends the data to the temporary storage means 2b for temporary storage. (Step ST3-2, step ST3-2
ST3-3).

The disk device control means 2c receives the command, sends the same command to the disk copy means 2e, interprets the command, and controls the disk device 4 (steps ST3-4 to ST3-6). Further, the disk device control means 2c takes out the data from the temporary storage means 2b and writes the data in the disk device 4 (step ST3-7,
ST3-8). This operation is performed for all necessary data, and the write operation is completed.

The disk device control means 2c sends the command and the data fetched from the temporary storage means 2b to the disk copy means 2e (steps ST3-9, ST3-).
10). The disk copy means 2e sends the command and data to the communication means 2d (step ST3-11). The communication means 2d sends the command and data to the communication means 5d via the sub bus 7 (step ST3-12). The communication means 5d sends the command and data to the disk copy means 5e (step ST3-13).

The disk copy means 5e sends a command to the disk device control means 5c and sends data to the temporary storage means 5b (step ST3-14). This temporary storage means 5b
Temporarily saves the data (step ST3-1).
5). This operation is performed for all necessary data, and the transmission operation is completed.

After that, the disk device control means 5c
Interprets the command to control the disk device 6, retrieves the data from the temporary storage means 5b based on the command, and writes the data in the disk device 6 (step ST3-16). Thus, by using the temporary storage means 2b, 5b, the data transmission time between the central processing unit 1 and the disk control units 2, 5 can be shortened, and the utilization efficiency of the central processing unit 1 can be further improved. .

Example 2. FIG. 4 is a flow chart for explaining a data write operation according to an embodiment of the invention described in claim 2. In the case of the embodiment 1 shown in FIG. 3, after writing all necessary data to the disk device 4, Sending commands and data to the controller 5,
In the case of this embodiment, the first and second disk control devices 2,
5 is provided with a simultaneous write function means, the first disk control device 4 writes data to the disk device 4 and commands the disk control device 5 via the disk copy means 2e and the communication means 2d. Then, the data is transmitted and the data is simultaneously written to the disk device 6 (steps ST4-1 to ST4-15). Therefore, it is possible to shorten the double data writing time for the disk devices 4 and 6.

On the other hand, when an abnormality is found in the disk controller 5 or the disk device 6, the disk controller 2 disconnects the sub-bus 7 so that the write request and the data are not sent to the disk controller 5 via the sub-bus 7. The disk controller 5 or disk device 6
Repair or replace.

For example, when an abnormality is found in the disk controller 2 or the disk controller 4, the central processing unit 1 changes so that all the operations are performed on the disk controller 5, and at the same time the disk controller 5 The sub bus 7 is disconnected so that write requests and data are not sent to the disk controller 2. At this time, the disk device 6 which has been a spare until now becomes the current one, and the central processing unit 1 makes a data write and read request to the disk controller 5 and the disk controller 2 or the disk device 4 in which the abnormality is found. Will be repaired or replaced.

When the replaced new disk control device 2 or disk device 4 is connected again, the active disk control device 5 or 6 connects the sub bus 7 and reserves from the active disk device 6 by utilizing the idle time. The data recovery operation is performed on the disk device 4 using the disk copy means 2e or 5e by the operation shown in the flowchart of FIG.

The recovery operation when the disk device 6 is replaced will be specifically described below with reference to the flowchart of FIG. First, the communication means 2a receives a disk copy command from the central processing unit 1 and sends the command to the disk copy means 2e via the disk device control means 2c (step ST5-1). Then, the disk copy means 2e instructs the disk device control means 2c to read the data of the disk device 4 (step ST5-2).

The disk device control means 2c reads data from the disk device 4 and stores this data in the temporary storage means 2
It is stored in b (step ST5-3). The disk copy means 2e reads the data from the temporary storage means 2b and sends it to the communication means 2d, and at the same time, the communication means 2d, the sub bus 7,
A write command is sent via the communication means 5a (step ST5-4).

The communication means 2d sends data to the communication means 5d (step ST5-5). The communication means 5d sends the data to the disk copy means 5e (step ST5-6).
The disk copy means 5e stores the data in the temporary storage means 5b and sends a write command to the disk device control means 5c (step ST5-7).

The disk device control means 5c reads data from the temporary storage means 5b (step ST5-8) and writes this data to the disk device 6 (step ST5).
-9). The disk copy means 2e judges whether or not all the contents of the disk device 4 have been copied, and if NO, returns to the step ST4-2 and repeats the above operation, and then Y
If it is ES, the operation ends (step ST5-1).
0).

Example 3. FIG. 6 is a block diagram showing an embodiment of the invention described in claim 4, and is a duplicated disk device 8
Is connected to a plurality of (two in the illustrated example) central processing units 1a and 1b via a system bus 3. Since the duplicated disk device 8 has exactly the same configuration as that of the first embodiment shown in FIG. 1, the same parts are designated by the same reference numerals and the duplicate description thereof will be omitted.

FIG. 7 is a flow chart for explaining the operation of this embodiment. Now, the central processing unit 1a makes a data read request from the disk device 4 to the duplicated disk device 8 (step ST7-1). The duplicated disk device 8 starts the data reading process from the disk device 4 (step ST7-2). During this process, the central processing unit 1
When b requests the redundant disk device 8 to write data to the disk device 6, the redundant disk device 8 notifies the central processing unit 1b that the processing is in progress and rejects the request (step ST7-3).

Data read processing from the disk device 4 of the duplicated disk device 8 ends (step ST7-4).
After that, the central processing unit 1b issues a data write request to the disk device 6 to the redundant disk device 8 (step S
T7-5). The duplicated disk device 8 starts the data writing process to the disk device 6 (step ST7-
6). When the central processing unit 1a issues a data read request from the disk device 6 during this processing, the duplicated disk device 8 notifies the central processing unit 1a that the processing is in progress and rejects the request (step ST7-7). .

Data writing processing to the disk device 6 by the duplicated disk device 8 is completed (step ST7-8).
After that, the central processing unit 1a issues a data read request from the disk device 6 to the duplicated disk device 8 (step S
T7-9) and the redundant disk device 8 is the disk device 6
To start the data reading process from (step ST7-
10) Then, the data reading process from the disk device 6 is completed (step ST7-11).

Thereafter, when the central processing unit 1b makes a data read request from the disk device 4 to the redundant disk device 8 (step ST7-12), the redundant disk device 8 starts the data read process from the disk device 4. (Step ST7-13) and the data reading process from the disk device 4 is completed in step ST7-14.

In this embodiment, the data stored in the disk devices 4 and 6 are stored in any of the central processing units 1a and 1b.
However, it can be used (read / write). Therefore, since a plurality of central processing units can share the data of one disk unit, centralized management of data is possible. Further, it is possible to eliminate the waste of having each central processing unit have data. The data write / read processing for the disk devices 4 and 6 is the same as the operation described in the first embodiment of FIG.

Example 4. FIG. 8 is a block diagram showing an embodiment of the invention described in claim 5, and the same parts as those in FIG. In FIG.
Reference numeral 11 denotes a duplication driver which is a part of the OS included in the central processing unit 1, and the duplication driver 11 is provided with software for executing the duplication writing process 12 and the data restoration process 13. Reference numeral 14 denotes a disk management table as a processing area management means provided in the central processing unit 1, and when the double writing process 12 or the data recovery process 13 is executed, the entire disk device 4 or 6 is divided into logical block units. It manages and stores the processing area for each logical block.

FIG. 9 shows the storage state of the disk management table 14 applied to the invention described in claim 6. For example, a plurality of sectors for each track are collectively set as a logical block unit. FIG. 10 is an image diagram of the data recovery processing 13, and either processing of pattern 1 or pattern 2 can be performed depending on how the disk management table 14 is used.

The operation of this embodiment will be described below.
First, the operation of pattern 1 will be described with reference to the flowcharts of FIGS. FIG. 11 shows a case where the disk device 6 is broken and replaced with a new disk device. After the operation is started, the contents of the disk management table 14 are first referred to, and it is judged whether the logical block to be restored is already copied ( Steps ST11-1, ST11-2), Y
If it is ES, the data to be copied is registered as the next several sectors (step ST11-3).

If the above judgment is NO, data of several sectors is read from the disk device 4 and written in the memory (which can be cached) 15 via the CPU 1-1 of the central processing unit 1 (step ST11-4). . Next, the data written in the memory 15 is read out via the CPU 1-1, written into the disk device 6 via the disk control device 5, and the disk management table 14 of the written data is updated (steps ST11-5 and ST11).
-6). Then, it is determined whether the data recovery is completed. If NO, the process returns to step ST11-1 to repeat the above operation, and if YES, the data recovery operation is completed (step ST11-7).

FIG. 12 shows the disk device 4 according to the pattern 1.
The processing will be described in the case where an interrupt occurs from the OS during the copying of the above data to the disk device 6 and writing from the OS to the disk device 4 occurs. After starting operation, first
The duplication driver 11 is notified that the OS has a write command in the disk device 4, and the write data memory 1
The start address and the data length on No. 5 are notified to the duplex driver 11 (step ST12-1).

The duplex driver 11 reads data from the memory 15 by the double writing process and writes this data in the disk device 4 (step ST12-2). The double writing process updates the disk management table 14 with respect to the written data, and then determines whether all the data has been written. If NO, the process returns to step ST12-2 to repeat the above operation, and if YES. If so, the operation ends (steps ST12-3 and ST12-4).

FIG. 13 is a flow chart for explaining the operation of the pattern 2, which is the case where the disk device 6 is broken and replaced with a new disk device. After the operation is started, the data of the next several sectors is read from the disk device 4 and written in the memory 15 (step ST13-1). The data written in the memory 15 is read from the memory and written in the disk device 6 (step ST13-2). After updating the contents of the disk management table 14 in response to this writing process, it is judged whether the data recovery process is completed (steps ST13-3 and ST13-4). Repeat the operation, YE
If S, the operation ends.

FIG. 14 shows the disk device 4 according to the pattern 2.
The processing will be described in the case where an interrupt occurs from the OS during the copying of the above data to the disk device 6 and writing from the OS to the disk device 4 occurs. After the operation starts, the OS informs the duplication driver 11 that there is a write command to the disk device 4, and informs the duplication driver 11 of the start address and the data length of the write data on the memory 15 (step ST14-1).

Next, by referring to the contents of the disk management table 14, it is judged whether or not the data to be read is already copied. If NO, the data to be written is registered as the next several sectors (steps ST14-2 to ST14-2). ST14
-4). If the above determination is YES, the duplication driver 11 reads the data from the memory 15 and writes this data to the disk device 4 by the duplication writing process (step ST14-5).

After that, the double writing process 12 updates the contents of the disk management table 14 in response to the writing process, judges whether or not all data writing is completed, and NO
If so, the process returns to step ST14-2 to repeat the above operation, and if YES, the operation ends (step ST1).
4-6, ST14-7).

Example 5. FIG. 15 is a block diagram showing an embodiment of the invention described in claim 7, and a register 16 is used instead of the disk management table 14 of the embodiment 4 shown in FIG.
Since the other configurations are the same as those of the fourth embodiment, the same portions are denoted by the same reference numerals and the duplicate description will be omitted.

FIG. 16 is a diagram showing the storage state of the register 16. For example, a flag indicating whether or not copying is in progress in area A, and the extent to which the disk drive has completed copying in areas B to D are shown in FIG. The indicated cylinder number, track number and sector number or logical block number are stored.

In this embodiment, the pattern 1 shown in FIG. 10 is used.
Cannot be processed. Therefore, the processing operation of pattern 2 will be described with reference to the flowcharts shown in FIGS. FIG. 17 shows a case where the disk device 6 is broken and replaced with a new disk device. After the operation starts, the disk device 4
The data of the next several sectors is read from the central processing unit 1
Write to the memory 15 via the CPU 1-1 (step ST17-1). The data written in the memory 15 is read out from the memory and written in the disk device 6 (step ST17-2). After updating the contents of the register 16 in response to the writing process, it is judged whether the data restoration process is completed (steps ST17-3 and ST17-4), and N
If it is O, it returns to step ST17-1 and repeats the above operation, and if it is YES, it ends the operation.

FIG. 18 shows the disk device 4 according to the pattern 2.
The processing will be described in the case where an interrupt is generated from the OS while the data is being copied to the disk device 6 and writing from the OS to the disk device 4 has occurred. After the operation starts, the OS informs the duplex driver 11 that there is a write command in the disk device 4, and informs the duplex driver 11 of the start address and the data length of the write data in the memory 15 (step ST18-1).

Next, by looking at the contents of the register 16, it is judged whether or not the data to be read is already copied. If NO, the data to be written is registered as the next several sectors (steps ST18-2 to ST18-4). ). If the above determination is YES, the duplex driver 11 reads data from the memory 15 and writes this data to the disk device 4 by the double writing process (step ST18).
-5). In this double writing process, the contents of the register 16 are referred to, if the sector number to be accessed is larger than the value held in the register 16, double writing is performed, and if it is smaller, only the normal disk device is written. The sector number has the maximum value at the end of the disk device.

Thereafter, it is judged whether or not all data writing is completed. If NO, the process returns to step ST14-2 to repeat the above operation, and if YES, the operation ends (step ST18-6).

Example 6. FIG. 19 is a block diagram showing an embodiment of the invention described in claims 8 and 10. The same parts as those of the embodiment 3 shown in FIG. In FIG. 19, reference numeral 17 indicates a duplex driver 1
The head position changing means is provided in a part of No. 1 and has a processing procedure for changing the head position of one of the disk devices when the corresponding track positions of the heads of the disk devices 4 and 6 are the same. Reference numeral 18 denotes a track information register provided in a part of the duplex driver 11 and stores the track position corresponding to the head of the disk device 4 or 6.

FIG. 20 shows the head 4 of the disk devices 4 and 6.
a and 6a show corresponding track positions. In the illustrated example, the disk devices 4 and 6 each have a track number N of one cylinder, and as shown in FIGS. In both devices, the heads 4a and 6a are located at the track 0 position. Therefore, in this state, when a read command (read command) is output from the central processing unit 1, the access speed is the same regardless of which disk device the data is read from. Therefore, it is necessary to change the head position of either one of the disk devices.

Example 7. FIG. 21 is a flow chart for explaining the head position changing operation according to the eleventh aspect of the present invention. In this operation, data writing processing is performed on the disk devices 4 and 6, and the heads of both disk devices 4 and 6 correspond to the corresponding tracks. When the positions are the same, the head position of one disk device is changed.

First, the head position changing means 17 determines whether both the disk devices 4 and 6 are operating normally (step ST21-1). If NO, the operation is ended and YES.
If so, it is determined whether or not the data writing process has been performed on the disk devices 4 and 6 (step ST21-2).

If the above judgment result is NO, the operation is ended, and if YES, the disk devices 4, 6 (HDD1, H
The track position i, j corresponding to the head of DD2) is read from the track information register 18 (step ST21).
-3), it is determined whether both track positions i and j are equal (step ST21-4).

If the judgment result is NO, the operation is terminated and Y
If ES, the head position changing means 17 is assumed to have tracks 0 to N of the disk device, and K = i + (N +
1) / 2, when K> N, j = K− (N + 1), K ≦
When N, a command for calculating the value of j such that j = K is issued to the central processing unit 1 to obtain the movement amount of the head 6a of the disk device 6. In the illustrated example, the head 4 a of the disk device 4
Since the track position of is 0, i = 0. Since the number of tracks N is 4, N = 4. Then, applying these values to the above equation, K = 2 (K
Is rounded down to the right of the decimal point), and since K ≦ N, j
= 2, that is, the head 6a of the disk device 6 is track 2
Need only be changed to the corresponding position, so the value of this j is written in the register 18 (step ST21-
5).

After that, the head position changing means 17 outputs a seek command to the disk device 6,
As shown in (c), the head 6a of the disk device 6 is moved to the track position j (step ST21-6).

By shifting the head position of the disk device 6 in this way, for example, when data read processing to the track 2 or the track 3 or the track 4 occurs next, the data is read from the disk device 6. As a result, the read speed becomes faster than when the head 6a is not moved. Further, even when data read processing is performed on the track 0 or the track 1, by reading the data from the disk device 4, it is possible to maintain the same speed as when the head is not moved.

In the sixth embodiment, the head position is changed after the data writing process is completed, but in the present embodiment, the duplex driver 11 sends a write command to the disk device 4 and the disk device 6 is sent after the command operation is completed. By utilizing the fact that the write command is sent in order, the head 4a of the disk device 4 that has completed the data writing process is changed to a position different from the head 6a of the disk device 6.

Example 8. FIG. 22A is a flow chart showing an embodiment of the invention described in claim 12. First, both the disk control devices 2 and 5 judge whether they are operating normally (step ST21-1a). When the operation is completed and the result is YES, the track position (i) of the head of the disk device 4 is read from the track information register 18 (step ST21-2a).

When it is assumed that the head position changing means 17 has tracks 0 to N of the disk device, K = i + (N + 1).
/ 2, the central processing unit 1 issues a command to calculate the value of j such that j = K- (N + 1) when K> N and j = K when K ≦ N.
Then, the movement amount of the head 6a of the disk device 6 is obtained. In the illustrated example, since the track position of the head 4a of the disk device 4 is track 0, i = 0. Also,
Since the number of tracks N is 4, N = 4. Therefore, if these values are applied to the above equation, K = 2 (K is rounded down after the decimal point). Since K ≦ N, j = 2,
That is, it is sufficient to change the head 6a of the disk device 6 to a position corresponding to the track position 2, so that j
Value is written in the register 18 (step ST21-3
a).

After that, the head position changing means 17 outputs a seek command to the disk device 6, and FIG.
As shown in (c), the head 6a of the disk device 6 is moved to the track position j (step ST21-4a).

Example 9. In all of the sixth and seventh embodiments, the amount of change in the head position is calculated by calculation.
In the invention described in 3, the heads 4a, 6 of the disk devices 4, 6 are
The position of a is previously determined to be different from each other, and when the writing of data is completed, the head 4a,
6a is moved.

FIG. 22B is a flow chart for explaining the operation of this embodiment. First, the duplex driver 11 judges whether or not the data write processing has been performed (step ST2).
2-1b), if NO, the operation is ended, and if YES, the head 4a is moved to the disk device 4 at a predetermined position ×
Give a command to move to XX. Upon receiving this command, the disk device 4 moves the head to the position XXX (step S
T22-2b).

Next, the head 6a is attached to the disk device 6.
Command to move to a predetermined position. The disk device 6 receiving this command moves the head to the position --- (step ST22-3b).

[0114]

As described above, according to the first aspect of the present invention, the disk copy means copies the data stored in one disk device to the other disk device via the communication means between the disk control devices. Since it is configured to do so, when performing the data transfer operation between the disk devices,
Occupancy of the central processing unit and the system bus can be eliminated, and the utilization efficiency and reliability of the central processing unit can be improved.

According to the second aspect of the invention, the disk controller writes the data to its own disk device.
Since the simultaneous writing function means is configured to write the data to the other disk device at the same time, the data double writing process can be speeded up.

According to the third aspect of the invention, the data written from the central processing means is stored in the temporary storage means, and the data writing to the disk device is performed by reading from the temporary storage means. The data transmission time from the central processing unit to the disk control unit can be shortened, and the utilization efficiency of the central processing unit can be further improved.

According to the invention of claim 4, the first and second
Since the disk control device of (1) is connected to a plurality of central processing units, each central processing unit shares one disk device, and centralized management of data is possible.

According to the fifth aspect of the present invention, the processing area in which the data double writing processing or the data restoration processing has been performed is recorded in the processing area management means, and the data double writing must be performed with reference to the recorded contents. For example, the data recovery process is performed, and if the recovery process is completed, double writing is performed.
The efficiency of data recovery processing can be improved.

According to the sixth aspect of the present invention, the disk management table is configured to hold the processing area of the disk device in the dual write processing and the data recovery processing for each logical block unit. By referring to the contents of the management table, it is not necessary to perform unnecessary data recovery processing or double writing processing.
Data recovery processing time can be shortened and efficiency can be improved.

According to the invention described in claim 7, since the current data recovery processing position of the disk device is held in the register, the data recovery processing can be efficiently performed by referring to the stored contents of this register. Since the disk device management can be performed and unnecessary double writing is not required, the data recovery processing time can be shortened and the efficiency can be improved.

According to the eighth aspect of the invention, when the track positions of the heads of the two disk devices are equal, the head position changing means changes the head position of one disk device to the head position of the other disk device. With this configuration, the seek time for moving the head to the data reading position is shortened, and the data reading speed can be increased. In particular, it is very effective when a data write process occurs and then a lot of read processes occur.

According to the ninth aspect of the present invention, the head position of the disk device in which the data has been written first is changed to a position different from the head position of the other disk device by the head position changing means. Therefore, the head position changing process can be efficiently performed.

According to the tenth aspect of the present invention, since the head of the disk device in which the writing of data has been completed is changed to a predetermined position by the head position changing means, the changed position of the head is determined. The head position changing process can be easily performed without the need.

According to the eleventh aspect of the invention, when the heads of the disk device in which the data has been written are located at the same track position, different track positions to which the heads should be moved are calculated, and the head positions are calculated at the track positions. Since the head is moved, it is possible to easily and reliably move the head of the disk device to different track positions.

According to the twelfth aspect of the present invention, since the track position to which the head of the disk device in which the data writing has been completed is to be moved is calculated and the head is moved to the track position, both disk devices can be moved. The head can be moved to different track positions quickly and reliably.

According to the thirteenth aspect of the invention, since the head of the disk device is moved to a predetermined track position after the writing of data is completed, it is extremely possible to move the head of the disk device to different track positions. It's easy to do.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the invention described in claim 1.

FIG. 2 is a flowchart illustrating a read operation according to the embodiment of FIG.

3 is a flowchart illustrating a data write operation according to the embodiment of FIG.

FIG. 4 is a flow chart illustrating a data read operation according to the embodiment of the invention described in claim 2;

FIG. 5 is a flowchart illustrating a data copy operation after the disk device is replaced.

FIG. 6 is a block diagram showing an embodiment of the invention described in claim 4;

FIG. 7 is a flowchart illustrating the operation of the embodiment of FIG.

FIG. 8 is a block diagram showing an embodiment of the invention described in claim 5;

FIG. 9 is a storage state diagram of a disk management table applied to the invention according to claim 6;

FIG. 10 is an image diagram showing a storage state of a disk device for explaining a data recovery processing operation.

FIG. 11 is a flow chart diagram for explaining the operation when the recovery process is not performed for the data double-written during the data recovery process when a new disk device is connected.

FIG. 12 is a flowchart illustrating an operation when an interrupt occurs during the data recovery process of FIG.

FIG. 13 is a flowchart diagram illustrating an operation of recovering all data at the time of data recovery processing when a new disk device is connected.

14 is a flow chart diagram illustrating an operation when an interrupt occurs during the data recovery process of FIG.

FIG. 15 is a block diagram showing an embodiment of the invention according to claim 7;

FIG. 16 is a storage state diagram of a register.

FIG. 17 is a flowchart diagram illustrating an operation of recovering all data during a data recovery process when a new disk device is connected.

FIG. 18 is a flow chart diagram illustrating an operation when an interrupt occurs during the data recovery processing of FIG.

FIG. 19 is a block diagram showing an embodiment of the invention described in claims 8 and 10.

FIG. 20 is an explanatory diagram showing track positions to which the head of the disk device corresponds.

FIG. 21 is a flow chart diagram for explaining a head position changing operation according to the invention of claim 11;

FIG. 22 is a flow chart illustrating a head position changing operation according to the embodiments of the invention described in claims 12 and 13;

FIG. 23 is a block diagram showing a conventional duplicated disk device.

FIG. 24 is a flowchart illustrating a data write operation by the device of FIG. 23.

FIG. 25 is a flowchart illustrating a data read operation by the device of FIG. 23.

FIG. 26 is a block diagram showing another conventional duplex disk device.

27 is a flowchart explaining a data write operation by the device of FIG. 26. FIG.

28 is a flowchart illustrating a data read operation by the device in FIG. 26.

[Explanation of symbols]

 1 central processing unit 2 disk control unit 2d communication means with other disk control unit 2e disk copy unit 4 disk unit 5 disk control unit 5d communication unit with other disk control unit 5e disk copy unit 6 disk unit 11 duplex driver 12 Double writing process 13 Data recovery process 14 Disk management table 16 Register 17 Head position changing means

Claims (13)

[Claims] 1. A duplicated disk device having first and second disk control devices which are connected to a central processing unit and store the same data written from the central processing unit in respective disk devices that control the same. , Each of the second disk control devices is provided with communication means between the disk control devices and disk copy means for copying the data stored in one disk device to the other disk device via this communication means. Redundant disk device. 2. The first and second disk control devices,
2. The simultaneous writing function means for writing the data to the other disk device at the same time when the data is written to the disk device controlled by each, through the disk copy means and the communication means. Redundant disk device. 3. The first and second disk control devices,
2. The duplicated disk device according to claim 1, further comprising a temporary storage means for temporarily storing the data written from the central processing unit. 4. The first and second disk control devices are connected to a plurality of central processing units.
The described redundant disk device. 5. A duplicated disk device having first and second disk controllers connected to a central processing unit and storing the same data written from the central processing unit in respective disk devices which control the central processing unit. The apparatus is a dual disk apparatus, comprising processing area management means for managing a processing area of the disk apparatus in the dual writing processing and the data recovery processing. 6. The disk management as the processing area management means, which manages the entire disk device in logical block units, and holds the processing area of the disk device in the dual writing process and the data recovery process for each logical block unit. 6. The dual disk device according to claim 5, wherein a table is used. 7. The processing area management means uses a register for holding the position of the disk device currently undergoing recovery processing during data recovery processing.
The described redundant disk device. 8. A duplicated disk device having first and second disk controllers connected to a central processing unit and storing the same data written from the central processing unit in respective disk devices that control the same. Head position changing means for changing the head position of one disk device to the head position of the other disk device when the track positions of the heads of the disk devices controlled by the respective second disk control devices are equal at the end of data writing. A dual disk device comprising: 9. The duplexer according to claim 8, wherein the head position changing means changes a head of the disk device, which has finished writing data, to a track position different from that of the head of the other disk device. Disk device. 10. The duplicated disk device according to claim 8, wherein the head position changing means changes the head of the disk device which has finished writing data to a predetermined track position. 11. It is determined whether or not the first and second disk control devices operate normally to write the data written by the central processing unit to the disk devices controlled by the first and second disk control devices, and write the data. If processing has been performed, the track positions corresponding to the respective heads of the disk device are read from the track information register and compared, and if the head positions of the respective disk devices are the same, the heads of one of the disk devices are equal. Is calculated, and the head is moved to the calculated track position. 12. It is determined whether or not the first and second disk control devices have normally operated to write the data written from the central processing unit to the disk devices controlled by the respective first and second disk control devices, and write the data. The track position corresponding to the head of the disk device whose processing has been completed first is read from the track information register, the track position where the head should be changed is calculated based on this track position, and the head is moved to the calculated track position. A method for changing the head position, which comprises: 13. It is determined whether the first and second disk control devices have normally operated, and the write processing of the data written from the central processing unit has been performed on the disk devices controlled by the first and second disk control devices. A head position changing method comprising moving each head of the disk device to a predetermined track position after the processing is completed.