JP2001100948A - Disk array device and its control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a disk array device and its control method which can automatically allocate a normal recording area where restored data can be rewritten. SOLUTION: The device is provided with an automatic allocation part 16 which automatically allocates a rewritable recording area in a disk drive where an error has occurred.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk array device and a control method therefor, and more particularly, to a method for reading data from an arbitrary disk device. The present invention relates to a redundant disk array device capable of restoring read data based on read data and a control method thereof.

[0002]

2. Description of the Related Art Conventionally, a plurality of disks are connected in parallel and connected to each other, and the whole of them is controlled as a single disk device. Disk array devices have been used as external storage devices for improving the performance.

In this disk array device, there are generally six types of levels from RAID 0 to RAID 5 depending on the control stage. Generally, RAID 3 to RAID 5 generate an error when an error occurs in reading data from the disk device. It has parity data for correction and is known as a so-called redundant disk array device. Further, among the disk array devices having such redundancy, RAIDs 3 and 4 use one of the disk devices as a disk device for parity recording.
Has a configuration in which parity data is allocated to each disk device without having a disk device for parity recording. In RAID 3, the recording area, which is the minimum unit for recording data, is a block area of 1 byte or 1 bit, whereas RAID 4 and 5 are sector areas of each sector.

As an example of such a redundant disk array device, a RAID 3 disk array device 1 shown in FIG. 3 is provided with a host computer 2 as a higher-level device that outputs a command such as data reading or writing.
The host computer 2 is connected to a RAID controller 4 via a host bus 3.

The RAID controller 4 includes a plurality of disk devices 5A, 5B on which read data is recorded.
And one parity recording disk device 6 is connected via a drive bus 7 in parallel with each other.

The data stored in each of the disk devices 5A and 5B is represented by "0" or "1" or a combination thereof, and the parity recording disk device 6 stores the data in each of the disk devices 5A and 5B. Exclusive OR of stored data is stored. For example, the uppermost block area b1 in FIG. 3 of the disk device 5A
"0" is stored in the corresponding disk device 5B.
If "1" is stored in the uppermost block area b1 'of the parity recording disk device 6, the "1" is stored in the uppermost block area b1 "of the parity recording disk device 6 corresponding to these block areas b1 and b1'. , "0" and "1" are stored as the exclusive OR 0XOR1 = "1".

[0007] The RAID control device 4 determines that the data in the block area b1 'of the disk device 5B is "
When a read error of “0” occurs, the data “0” of the corresponding block area b1 of another disk device 5A.
Exclusive OR 0XOR1 of the data "1" of the corresponding block area b1 "of the parity recording disk device 6
= "1" restores the data "1" in the block area b1 'of the disk device 5B in which the read error has occurred.

Further, the RAID controller 4 outputs the restored data to the host computer 2 and rewrites the restored data to a block area other than the failed block area in the disk device 5B. Has become.

As a result, the reading error can be properly corrected, and the disk array device 1 having high reliability and high functionality can be provided.

[0010]

However, in the conventional disk array device 1, when rewriting the restored data, if a new failure occurs in the recording area to be rewritten, the data cannot be rewritten. For example, problems such as lack of system reliability have occurred.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and an object of the present invention is to provide a disk array device capable of automatically allocating a normal recording area in which restored data can be rewritten, and a control method thereof. It is the purpose.

[0012]

According to a first aspect of the present invention, there is provided a disk array apparatus comprising:
An automatic allocating unit for automatically allocating a rewritable recording area in a disk device in which an error has occurred is provided.

By adopting such a configuration, a rewritable normal recording area can be automatically allocated, and the reliability of the system can be improved.

A feature of the disk array device according to the second aspect is that the disk array device according to the first aspect has control means for automatically restoring the data in which the error has occurred and rewriting the restored data.

[0015] By adopting such a configuration, the restoration of the data in which the reading error has occurred and the rewriting of the data can be performed quickly, so that the speed and reliability of the system can be further improved. Can be.

A feature of the control method of the disk array device according to the third aspect is that a rewritable recording area in the disk device in which an error has occurred is automatically allocated.

By employing such a method, a rewritable normal recording area can be automatically allocated, and the reliability of the disk array device can be improved.

A feature of the control method of the disk array device according to the fourth aspect is that, in the third aspect, restoration of the data in which the error has occurred and rewriting of the restored data are automatically performed.

By adopting such a configuration, it is possible to quickly restore data in which a read error has occurred and rewrite the data, thereby further improving the speed and reliability of the system. Can be.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a disk array device according to the present invention will be described below with reference to FIGS.

As shown in FIG. 1, the disk array device 10 according to the present embodiment has a host computer 11 as a higher-level device that outputs a command such as data reading or writing.
1 has an RA as a control means via the host bus 12.
An ID control device 13 is connected, and the RAID control device 13 performs various controls necessary for reading, writing, and restoring data.

The RAID controller 13 includes, for example,
A plurality of disk devices 14 such as a SCSI interface type hard disk drive are connected in parallel via a drive bus 15. In each of these disk devices 14, read data is divided and recorded for each fixed recording area. This recording area is RA
In the case of ID3, it is a block area of one bit or one byte, and in the case of RAID4 and RAID5, it is a sector area of one sector. Also, RAID3 and RAI
In the case of D4, it is necessary that at least one of the plurality of disk devices 14 is a disk device for parity recording in which parity data is recorded. In the case of RAID 5, parity data is stored in each disk device 14. What is necessary is just to sort and store. In the case of RAID1, so-called mirroring, at least one pair of disk devices 14 on which the same data is recorded is provided, so that parity data is not required.

When a read error occurs when reading data from a recording area of an arbitrary disk device 14, the RAID control device 13 transmits data in a corresponding recording area of another disk device 14 and, if necessary, The errored data is automatically restored based on the corresponding parity data, and the restored data is automatically rewritten into a normal recording area of the disk device 14 in which the error has occurred. ing.

Further, the RAID control device 13
An automatic allocating unit for automatically allocating a normal rewritable recording area when rewriting the data by the ID control unit; Other recording areas are automatically assigned.

Next, an embodiment of a method for controlling a disk array device according to the present invention to which the disk array device 10 is applied will be described.

First, as shown in step 1 (ST1), the disk device 14 is
Outputs a command to read data in Next, in step 2 (ST2), it is determined whether or not an error has occurred in reading data by the RAID control device 13. If it is determined that no error has occurred, the control for the error is terminated as shown in step 3 (ST3), and the data in the disk device 14 is read as it is.

On the other hand, if it is determined that an error has occurred, the process proceeds to step 4 (ST4), and it is determined whether or not the disk array device 10 is RAID0.

In this step 4, if the data is RAID0, the data cannot be restored.
After returning an error as shown in 5), step 3 (ST
Proceed to 3) and end the control. On the other hand, if it is not RAID 0, the process proceeds to step 6 (ST6), and it is determined whether or not the mode is the degeneration mode, that is, whether or not one disk device 14 has failed as a whole.

In step 6 (ST6), if the mode is the degeneration mode, restoration is impossible, and
After proceeding to 5) and returning an error, step 3 (ST3)
To end the control. On the other hand, if the mode is not the degeneration mode, the process proceeds to step 7 (ST7), and it is determined whether or not there is a media error, that is, whether or not a failure occurs only in an arbitrary recording area of one disk device 14.

If it is determined in step 7 (ST7) that the error is not a media error, the data cannot be restored, so that the process proceeds to step 8 (ST8) to make a failed disk, and the process proceeds to step 9 (ST9). It proceeds and recognizes that it is in the degeneration mode. Then, step 1 (ST
After returning to 1), steps 2, 4, 6, 5 (ST2,
The process proceeds to step 3 (ST3) via (4, 6, 5), and the control is terminated. On the other hand, if it is determined that the error is a media error, the process proceeds to step 10 (ST10) to read data from another disk device 14 in which the error has occurred.

The data to be read at this time is RAID1
Is the same data read from another disk device 14 on which exactly the same data as the disk device 14 in which the error occurred is recorded. In the case of RAIDs 3 and 4, the data (parity) in the block (sector) area corresponding to the block (sector) area of the other disk device 14 where the error has occurred and the corresponding block (not shown) of the disk device (not shown) This is parity data in a (sector) area. In the case of RAID 5, it is data in a block (sector) area corresponding to the block (sector) area of the other disk device 14 where the error has occurred, and parity data allocated to each of the other disk devices 14.

Then, as shown in step 11 (ST11), it is determined whether or not an error has occurred when reading data from another disk device 14.

In step 11 (ST11),
If it is determined that an error has occurred, since data cannot be restored, the process proceeds to step 8 (ST8) and a failed disk is set. On the other hand, if it is determined that no error occurs, the process proceeds to step 12 (ST12) and the R
The other disk device 14 is controlled by the AID controller 13.
Automatically restores data with errors based on the data read from.

After the restoration of the data, the automatic allocating unit 16 of the RAID control unit 13 sets the restored data as a rewrite area for rewriting the restored data to the failed disk device 14. Of the fourteen recording areas, a normal recording area other than the failed recording area is automatically allocated.

Then, as shown in step 14 (ST14), it is determined whether an error has occurred in the allocated recording area, that is, whether the allocated recording area is rewritable.

In step 14 (ST14),
If it is determined that an error has occurred, it is determined that rewriting cannot be performed, and the process proceeds to step 8 (ST8) to designate a failed disk. On the other hand, if it is determined that no error occurs, it is determined that rewriting is possible, and the process proceeds to step 15.
The restored data is automatically rewritten in this recording area.

Therefore, in steps 13 (ST13) and 14 (ST14), a rewritable recording area is automatically allocated.

Further, proceeding to step 16 (ST16), if an error occurs at the time of rewriting, proceeding to step 8 (ST8) to determine a failed disk. On the other hand, if no error occurs, the process returns to step 1 (ST1), and the rewritten data is read again by the host computer 11. Then, steps 1 and 2 (ST1,
If no error occurs after 2), step 3 (ST)
Proceed to 3) and end the control.

Therefore, according to the present embodiment, since the rewritable recording area can be automatically allocated by the automatic allocating unit 16, the restored data can be properly allocated to a normal recording area other than the failed recording area. Can be rewritten, and the reliability of the system can be improved.

Further, restoration of data in which a read error has occurred and rewriting of the restored data are performed by the RA.
Since the data can be automatically performed by the ID control device 13, the data can be restored and rewritten quickly, and the speed and reliability of the system can be further improved.

The present invention is not limited to the above-described embodiment, but can be variously modified as needed.

[0042]

As described above, according to the disk array device of the first aspect and the control method of the disk array device of the third aspect of the present invention, the restored data can be normally recorded in areas other than the failed recording area. Rewriting can be appropriately performed in the area, and the reliability of the system can be improved.

According to the disk array device of the second aspect and the control method of the disk array device of the fourth aspect, it is possible to quickly restore data having a read error and rewrite the restored data. Therefore, in addition to the effects of the disk array device according to the first aspect and the control method of the disk array device according to the third aspect, the speed and reliability of the system can be further improved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a disk array device according to the present invention.

FIG. 2 is a flowchart illustrating an embodiment of a method for controlling a disk array device according to the present invention.

FIG. 3 is a block diagram showing a conventional disk array device.

[Explanation of symbols]

 Reference Signs List 10 disk array device 11 host computer 13 RAID controller 14 disk device 16 automatic allocation unit

Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat II (reference) G11B 19/02 501 G11B 19/02 501A 20/10 20/10 C 20/12 20/12 (72) Inventor Shigeru Araki Alps Electric Co., Ltd. (72) Inventor Mitsuo Goto 1-7 Yukitani Otsukacho, Ota-ku, Tokyo Alps Electric Co., Ltd. (72) Yuichiro Sawame Tokyo, Ota-ku, Tokyo 1-7, Yutsuya Otsuka-cho, Ota-ku Alps Electric Co., Ltd. (72) Inventor Hiroyoshi Usui 1-7, Yukitani Otsuka-cho, Ota-ku, Tokyo Alps Electric Co., Ltd. (72) Inventor Takechiyo Takatsuki Tokyo 1-7 Yukitani Otsukacho, Ota-ku Alps Electric Co., Ltd. (72) Inventor Kentaro Inoue 1-7 Yukitani Otsukacho, Ota-ku, Tokyo Alps Electric Co., Ltd. (72) Inventor Mitsuru Maruyama Ota-ku, Tokyo 1-7, Yutani-Otsuka-cho Alps Electric Co., Ltd. F-term (reference) 5B018 GA06 HA12 HA35 KA12 KA13 KA15 KA22 MA14 QA14 QA15 RA11 5B065 BA01 CA30 EA03 EA15 EA24 5D044 BC01 CC04 DE62 DE64 DE66 DE68 HL01 5D066 AA02 AA08 BA02 BA07

Claims (4)

[Claims] 1. A disk drive in which a plurality of disk drives are connected via a parallel interface and an unrecoverable error occurs when reading data from a recording area of any of the disk drives. Restoring the data in which the error has occurred based on the data in the corresponding recording area of the disk device other than the above, rewriting the restored data to the disk device in which the error has occurred, and transferring the restored data to the host device A disk array device comprising an automatic allocating unit that automatically allocates a rewritable recording area of the disk device in which the error has occurred. 2. The disk array device according to claim 1, further comprising control means for automatically restoring the data in which the error has occurred and rewriting the restored data. 3. A disk on which an unrecoverable error occurs when reading data from a recording area of an arbitrary disk device among a plurality of disk devices connected via a parallel interface. Restoring the data in which the error has occurred based on the data in the corresponding recording area of the disk device other than the device, rewriting the restored data to the disk device in which the error has occurred, and placing the restored data in the upper What is claimed is: 1. A method for controlling a disk array device, which controls output to a device, wherein a rewritable recording area of the disk device in which the error has occurred is automatically allocated. 4. The method according to claim 3, wherein the restoration of the data in which the error has occurred and the rewriting of the restored data are automatically performed.