JP2000293315A - Disk array device and spare disk rotation method - Google Patents

Abstract

(57) [Summary] [Problem] To improve the reliability by averaging the life of HDDs and
Provided is a disk array device that reduces the number of replacements of DD. The power-on time of HDDs 10-1 to 10-5 is managed by firmware, and the HDD with the longest power-on time among the HDDs 10-1 to 10-5 is rotated to a hot spare disk to reduce the power-on time as much as possible. Since a short HDD can be used as a data disk, the reliability of the disk array is improved and the number of replacements of the HDD is reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a disk array device used in a computer system.

[0002]

2. Description of the Related Art Conventionally, as a storage device in a large-scale computer system with a strong demand for high reliability together with a large storage capacity, a redundant RAID (Redundant Arrays) has been used.
f Independent Disks) have been developed. Even if the data of one HDD is lost due to some kind of failure among the plurality of hard disks (HDDs) constituting the disk array, the data of the remaining HDDs constituting the disk array is And you can restore lost data by RAID function to restore data from parity data. However, in a disk array device, if only one HDD has lost data, data can be restored using the RAID data restoration function. However, if data from two or more HDDs is lost, the RAID data restoration function can be used. However, there was a disadvantage that the data could not be restored. Therefore, a hot spare disk is installed in the disk array device in advance as a backup HDD in the event of a failure, and one HDD
When the data is lost, the data of the HDD is restored by the data restoration function of the RAID, and the function of automatically saving the restored data to the backup HDD is provided.

[0003] By doing so, the reliability of the disk array device can be improved.
In the long term, the life of the device is not always sufficient, and after a certain period of time, the reliability may be sharply reduced. Since the warranty life of each HDD is usually about 5 years, reliability problems will occur after 5 years. Actually, since the life of each HDD varies, after the warranty period has expired, failures occur one after another due to wear of components constituting the drive. Therefore, even if a RAID is configured by combining multiple HDDs to obtain high reliability, after five years or more throughout the year, more than half of the HDDs will statistically have a lifetime and HDD failures will occur one after another. , HDD
Need to be replaced.

[0004]

The life of the HDD is determined by the other components (the controller 10) constituting the disk array device.
Year, power supply 7 years), so the number of replacements is increased. Further, since the price of the HDD and the work cost due to the movement of the data are also expensive, a device for minimizing the replacement of the HDD is required as much as possible. An object of the present invention is to provide a disk array device that averages the life of the HDD, improves the reliability of the HDD, and reduces the number of replacements of the HDD by rotating the HDD serving as a hot spare disk in the disk array device. And

[0005]

According to the present invention, in order to achieve the above object, in a disk array device having a plurality of storage media by a disk array controller, at least one of the plurality of storage media is defined as a spare disk. And a means for rotating a storage medium defined as the spare disk.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows a block diagram of a disk array device. 1 is a host computer, 2 is a disk array device, 3 is a disk array controller that controls each HDD, and 4 is a host input / output unit that exchanges data between the host computer 1 and the disk array controller 3 and temporarily stores data. 5, a microprocessor (MPU) for controlling the entire disk array controller 3; 6, a memory storing firmware which is a program for operating the MPU; 7, a data buffer acting as a buffer for the disk array controller 3; The system in the array controller 3
A bus 9, an HDD input / output unit for exchanging data between the disk array controller 3 and the HDD and temporarily storing the data; 10-1 to 10-5, HDDs for storing data;
D. In this figure, 10-4 is allocated to a parity disk storing parity data, and 10-5 is allocated to a hot spare disk of a backup HDD in the event of a failure. Next, FIG. 2 shows a flow of an operation when the hot spare disk is periodically rotated. FIG.
8 shows the transition of the HDDs 10-1 to 10-5 and the data and parity data when the hot spare disk is rotated periodically. Microprocessor 5
The parity data stored in the HDD 10-4 is stored in the HDD 1
An instruction to write to 0-5 is issued (step 21). HDD10
-4 parity data is temporarily stored in the data buffer 7 (step 22). Write the parity data stored in the data buffer 7 to the HDD 10-5 (step 2).
3). The microprocessor 5 confirms that the writing of the parity data is completed (step 27),
0-4 is redefined from the parity disk to the hot spare disk, and the HDD 10-5 is redefined from the hot spare disk to the parity disk (step 28).

FIG. 3 shows a data flow when a write request is issued from the host computer 1 while the parity data stored in the HDD 10-4 is being written to the HDD 10-5. A write request is received from the host computer 1 and the write data is temporarily stored in the data buffer 7.
(Step 25). The data stored in the data buffer 7 is written to both the HDD 10-4 and the HDD 10-5 (Step 26). This is to make it possible to return to the original state 1 in FIG. 2 when writing to the HDD 10-5 fails due to a write request from the host computer 1. The same process is repeated below.
The HDD that will be the hot spare disk is rotated regularly. By regularly rotating the HDD that will be the hot spare disk, it can also serve as a regular patrol of the hot spare disk.
A failure of the hot spare disk can be found at an early stage, and there is an effect of improving the reliability of the disk array device. next,
A second embodiment will be described. If the firmware 6 in the disk array controller 3 is HDD 10-1 to 10-
5 to control the power-on time, and
The HDD having the longest energization time among the HDDs 10-1 to 10-5 is rotated so as to be a hot spare disk.

The firmware in the memory 6 stores information for managing the power-on time of the HDDs 10-1 to 10-5 in all the HDDs in the management table shown in FIG. The firmware periodically sets the energization time of the HDDs 10-1 to 10-5 (approximately 1
(Period every minute) and adds to the total energizing time in the management table to manage the energizing time, and the longest energizing time
Recognize the HDD. When the configuration of the disk array changes due to a failure of the HDD or the like, the information of the power-on time is read from the management table, and the disk array configuration is changed so that the HDD with the longest power-on time becomes a hot spare disk. For example, FIG.
0-2 shows a change in the configuration of the disk array when a failure occurs and a new HDD is replaced. State 1 is a normal operation state, in which HDDs 10-1 to 10-3 are defined as data disks for storing data, HDD 10-4 as a parity disk for storing parity data, and HDD 10-5 as a hot spare disk. . When the HDD 10-2 breaks down, the data stored in the HDD 10-2 restores the data from the data and parity data in another HDD. The restored data was a hot spare disk
It is immediately stored in the HDD 10-5 (state 2). The HDD 10-5 in which the data is stored becomes a data disk, and the HDD 10-
2 is replaced with a new HDD as a failed HDD and becomes a hot spare disk (state 3). The firmware uses the HDD with the longest power-on time for the hot spare disk and the HDD with the shortest power-on time for the data disk.
The data stored in the HDD 10-1 is replaced with a new one and written to the HDD 10-2 having a short energization time (state 4). HDD
After the data writing to 10-2 is completed, the HDD 10-1 is defined as a hot spare disk, the HDD 10-2 is defined as a data disk, and returns to the normal operation state (state 5).

As described above, when the firmware is stored in the HDD 10-1
It manages the power-on time of the HDD 10-1 to 10-5.
By rotating the HDD with the longest energization time among the hot spare disks among the HDDs in No. 5, the HDD with the shortest energization time can be used as a data disk, thereby improving the reliability of the disk array.

[0010]

According to the present invention, by rotating a storage medium defined as a spare disk, the reliability of the disk array device can be improved and the number of HDD replacements can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram of a disk array device.

FIG. 2 is a diagram showing an operation in the first embodiment of the present invention.

FIG. 3 is a diagram showing a state change of the HDD according to the first embodiment of the present invention.

FIG. 4 is a diagram showing a flow of data during rotation.

FIG. 5 is a view showing a management table of energization time of the HDD.

FIG. 6 is a diagram showing a state change of an HDD according to a second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Host computer 2 ... Disk array device 3 ... Disk array controller 4 ... Host input / output unit 5 ... Microprocessor (MPU) 6 ... Memory 7 ... Data buffer 8 ... System bus 9 ... Hard disk (HDD) input / output unit 10- 1-5: Hard disk (HDD)

Claims (3)

[Claims] In a disk array device having a plurality of storage media in a disk array controller, means for defining at least one of the plurality of storage media as a spare disk, and means for rotating the storage medium defined as the spare disk A disk array device comprising: 2. The apparatus according to claim 1, further comprising means for managing the power supply time of said plurality of storage media, and means for defining a storage medium having the longest power supply time as a spare disk.
The disk array device according to the above. 3. In a disk array device having a plurality of storage media by a disk array controller, a step of managing energization time of the plurality of storage media, and defining data of the storage medium having the longest energization time as the spare disk. And writing the storage medium having the longest energization time as a spare disk.