JP3454183B2 - Disk array device - Google Patents

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 more particularly to a technique for constructing a disk array using disk devices having different capacities.

[0002]

2. Description of the Related Art A disk array device for distributing and storing data in a plurality of disk devices depends on how the data is distributed and stored.
Although it is classified into a configuration such as D5, it is common that all are constructed by using disk devices of the same capacity. Although it is possible to configure a disk array by using disk devices of different capacities, in that case, even if the capacity of each disk device is large, the smallest capacity disk in the disk array is configured. It was used only as the same capacity as the device, and the usage efficiency of the disk capacity was significantly reduced.

In view of the above circumstances, when a disk array is constructed by using disk devices having different capacities, a technique for enabling a disk area which cannot be used as a conventional disk array to be used for the disk array is disclosed in Japanese Patent Laid-Open No. Hei 10-199903 There is a technique described in JP-A 8-249132.
In this technology (hereinafter referred to as "conventional technology"), one disk array is constructed by using the disk area from the beginning of the disk area in each disk device to the disk area of the same capacity as the minimum capacity disk apparatus, and disks other than the minimum capacity are configured. The extra disk area in the device is used for another disk array. For example, four 1 Gbyte disk devices and 50
If there is one 0 MB disk device, 1G
A disk array of 500 M x 5 = 2.5 Gbytes is configured by using the first 500 Mbytes of the disk unit of the byte disk unit and the entire disk region of the 500 Mbytes of disk unit, and the remaining 50 of the 1 Gbyte disk unit.
500Mx4 = 2 using 0MB disk area
Configure another disk array of G bytes. A similar technique is also described in JP-A-9-120342.

[0004]

According to the above-mentioned conventional technique, the disk area which could not be used conventionally as the disk array can be used for the disk array, and the usage efficiency of the disk area can be improved. However, the maximum capacity of one disk array that can be constructed is limited to “minimum capacity disk capacity × number of disk devices”, and it is not possible to realize the large capacity that is the greatest feature of the disk array.

Therefore, an object of the present invention is to make it possible to construct a disk array having a capacity which is the same as or approximately the same as the total capacity of all disk devices, by using disk devices having different capacities.

[0006]

In order to achieve the above-mentioned object, the present invention is achieved when a plurality of real disk devices in which at least any two real disk devices have different capacities are divided into disk areas of the same capacity. Each virtual disk device is used as a virtual disk device, and one of the virtual disk devices of the same capacity is used as a parity disk, including multiple virtual disk devices created on the same real disk device.
Using all virtual disk devices of
One disk array corresponding to RAID 3 having a capacity that is equal to or close to the total capacity of the plurality of real disk devices is configured. More specifically, in a disk array device composed of a plurality of real disk devices in which at least any two real disk devices have different capacities and a control device connected to the plurality of real disk devices,
The control device has a size register that holds a size for dividing the disk areas of the plurality of real disk devices, and an individual disk area obtained by dividing the plurality of real disk devices by the size as a virtual disk device, and each virtual disk One of the plurality of virtual disk devices including a means for generating a disk configuration table indicating the correspondence between the device and the real disk device, and a plurality of virtual disk devices generated on the same real disk device is a parity disk. Used for
Using all virtual disk devices of
Means for configuring one disk array corresponding to RAID 3 having a capacity that is the same as or approximately the same as the total capacity of the plurality of real disk devices, and the disk configuration table when a read and write request is made from the host device to the disk array And means for converting the address for the virtual disk device into the corresponding address for the real disk device.

[0007]

[0008]

Further, according to the present invention, a virtual disk device in a real disk device having the smallest number of generated virtual disk devices is a parity disk, and a virtual disk device created on another real disk device is a data disk. A disk array corresponding to RAID3 is configured by using the disk array.

Further, the present invention further comprises means for obtaining the greatest common divisor of the capacities of the plurality of real disk devices, and the value obtained by the means is set in the size register.

Further, the present invention further comprises a size setting register for holding a minimum size, and means for obtaining the greatest common divisor of the capacities of the plurality of real disk devices, and the value obtained by the means is the minimum size. If it becomes smaller, the minimum size is set in the size register, and otherwise, the obtained value is set in the size register.

In the disk array device of the present invention,
Multiple virtual disks of the same capacity, including multiple virtual disk devices created on the same real disk device, are the individual disk areas when dividing multiple real disk devices into disk areas of the same capacity. Since a disk array is configured using devices, RAID3 with a capacity that is the same as or close to the total capacity of all disk devices.
It is possible to construct a disk array corresponding to .

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION Next, an example of an embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a diagram for explaining the principle of an embodiment of the present invention. Now, consider a case where a disk array is constructed using two real disk devices D1 and D2 having different capacities as shown in FIG. Here, it is assumed that the capacity of the real disk device D1 is 1.5 times that of the real disk device D2. Conventionally, one disk array is configured by the entire disk area of the real disk device D2 having a small capacity and the disk area corresponding to the capacity of the real disk device D2 from the head of the real disk device D1. The rest of the disk space was unused. JP-A-8-249
In Japanese Patent Laid-Open No. 132, the remaining disk area of the unused real disk device D1 is utilized as another disk array. Therefore, although the use efficiency of the disk area is improved, two disk arrays, that is, a disk array having twice the capacity of the real disk device D2 and a disk array having half the capacity of the real disk device D2 have been formed.

In the case of the present invention, as shown in FIG. 1B, the disk areas of the real disk devices D1 and D2 are divided into, for example, the size of the greatest common divisor of each capacity, and each divided area is virtual. Disk devices D11 to D13, D21, D
22 and these virtual disk devices D11 to D13,
Each of D21 and D22 is regarded as one real disk device to form one disk array. As a result, the capacity of the disk array becomes equal to the total value of the capacities of the real disk devices D1 and D2, and a large capacity disk array can be constructed. Hereinafter, examples of the embodiment of the present invention will be described.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Referring to FIG. 2, one embodiment of a disk array device of the present invention is a disk array controller (DAC) 10 connected to a CPU 1 which is a host device, an address line 12 and It is composed of a plurality of real disk devices Disk1 to DiskN connected through the data line 13. Here, at least any two of the plurality of real disk devices Disk1 to DiskN have different capacities.

Further, the disk array controller 10
Includes a control unit 3, an interface 2 connected between the control unit 3 and the CPU 1, and a virtual disk configuration unit 11 connected to the control unit 3.

The virtual disk construction unit 11 further includes an address conversion unit 4 and a size calculation unit 6 connected to the control unit 3.
And a size register 5 provided between the address conversion unit 4 and the size calculation unit 6, and the address conversion unit 4 incorporates a disk configuration table 7.

Referring to FIG. 3, one embodiment of the controller 3 includes an MPU 301 connected to the interface 2.
A RAM 303 holding a program to be executed by the MPU 301 and various kinds of information, a cache 302 connected to the MPU 301, a selector 304, interfaces 306 and 307, and a selector 304 and the real disk devices Disk1 to DiskN are provided. Disk interfaces 305-1 to 305-N, and are connected to the address conversion unit 4 via the interface 306 and to the size calculation unit 6 via the interface 307.

FIG. 4 is a flow chart showing an example of a procedure for constructing a disk array using a plurality of real disk devices. First, the actual disk device used for the disk array is determined (step S1). This is done by the system administrator and by notifying the MPU 301 in the control unit 3 of the disk array controller 10 via the CPU 1. Further, when all the real disk devices actually connected to the disk array controller 10 are used for the disk array, the real disk device in which the MPU 301 is mounted via the selector 304 and the disk interfaces 305-1 to 305-N. It is also possible to automatically determine by recognizing. In the following, it is assumed that one disk array is constructed by using all the real disk devices Disk1 to DiskN.

Next, the number (ID) and capacity of each real disk device used for the disk array is acquired and temporarily stored in the RAM 303 (step S2). This is because the MPU 301 of the control unit 3 makes each of the real disk devices Disk1 to Disk
The N number and capacity may be acquired and stored in the RAM 303, or may be stored in the RAM 303 in the control unit 3 of the disk array controller 10 via the CPU 1 by the system administrator himself.

Next, the capacity (size) of the virtual disk device
Is determined (step S3). This is done by reading the capacity of each real disk device used for the disk array from the RAM 303 and giving it from the control section 3 to the size calculation section 6 through the interface 307 so that the size calculation section 6 can determine it. In the case of the present embodiment, the size calculation unit 6 obtains the greatest common divisor of the capacity of each given real disk device, determines the value as the size of the virtual disk device, notifies the MPU 301 and stores it in the size register 5. Let

Next, the MPU 301 of the control unit 3 uses the notified size for each of the physical disk devices Disk1 to DiskN.
Each of the divided areas obtained by dividing the disk area of 1 is set as one virtual disk device, and a disk configuration table indicating the correspondence relationship between the virtual disk device and the real disk device is created (step S4). FIG. 5 shows an example of the disk configuration table. In the disk configuration table, for each virtual disk device, a real disk device number indicating on which real disk device it is created, and a serial number indicating the number of the virtual disk device of that real disk device. Is recorded. The serial numbers are numbered 1, 2 ... In order from the head of the disk area of the actual disk device. For example, as shown in FIG. 6, in the case of the virtual disk device DIN corresponding to the disk area from the beginning of a certain real disk device to (N−1) × size to N × size, the serial number is N. The disk configuration table created as described above is stored in the RAM 303 of the control unit 3 and is also sent to the address conversion unit 4 via the interface 306 and held in the address conversion unit 4 as the disk configuration table 7. .

Next, in order to construct one disk array using the plurality of virtual disk devices generated as described above, the disk array configuration information is created and the RAM 30 is created.
3 (step S5). This process is almost the same as the process for constructing a disk array by a real disk device, except that a virtual disk device is used instead of the real disk device. In this embodiment, the disk arrays that can be constructed are RAID0, RAID1, and R.
There are three types of AID3. Which is to be created is specified in the MPU 301 by the system administrator through the CPU 1, for example. Of course, when the type of RAID to be created is designated in advance by the control parameter stored in the RAM 303, the designation by the system administrator is unnecessary. Less than,
A specific configuration example of the disk array will be described for each RAID.

For convenience of explanation, there are three real disk devices D1, D2 and D3 used for the disk array, the real disk device D3 has the smallest capacity, and the real disk device D2 has twice the real disk device D3. It is assumed that the real disk device D1 has a capacity, and the real disk device D1 has a capacity twice that of the real disk device D2. Further, the size calculating unit 6 determines the same value as the capacity of the real disk device D3 as the greatest common divisor, so that the real disk device D1 is divided into four to generate four virtual disk devices D11, D12, D13, D14. Then, the real disk device D2 is divided into two to generate two virtual disk devices D21 and D22.
It is assumed that 3 is not divided and one virtual disk device D31 is generated.

(1) RAID0 In the case of RAID0, as shown in FIG. 7, four virtual disk devices D11 to D14 on the real disk device D1 and two virtual disk devices D21 on the real disk device D2,
A disk array is constructed using all seven virtual disk devices, that is, D22 and one virtual disk device D31 on the real disk device D3. Then, logical addresses are assigned to the disk area of each virtual disk device as shown in 1 to n of FIG. 7, and as the disk array configuration information, for example, as shown in FIG. 8, for each logical address, Information that records the number of the virtual disk device in which the area pointed to by the logical address exists and the disk address (offset address from the beginning) is generated. Note that S is the size of the virtual disk device, and L is the address increment value per logical address.

(2) RAID1 In the case of RAID1, since a virtual disk device that is a master and a virtual disk device that is a slave are paired, an even number of virtual disk devices are used. The master virtual disk device and the slave virtual disk device do not exist on the same real disk device. A disk array corresponding to RAID 1 is configured by using as many virtual disk devices as possible within the range that satisfies these conditions. In the present example, as shown in FIG. 9, six virtual disk devices D1 other than the virtual disk device D14 are used.
1 to D13, D21, D22, D31 are used to construct a disk array, and D11 and D21, D12 and D31, D
13 and D22 are paired respectively. Then, logical addresses as shown in 1 to n of FIG. 9 are applied to the disk areas of the respective virtual disk devices D11 to D13, D21, D22, D31, and as disk array configuration information, for example, shown in FIG. As described above, for each logical address, the information recording the number of the virtual disk device serving as the master, the number of the virtual disk device serving as the slave, and the in-disk address (offset address from the head) is generated.

When the number of generated virtual disk devices is an odd number, the system administrator determines through the CPU 1 which virtual disk device is unused.
It may be specified as U301, or the MPU 301 itself may make the last virtual disk device (the one farthest from the head of the disk area) of the real disk device in which the maximum number of virtual disk devices have been created unused. It may be done. Furthermore, the virtual disk device to be paired may be designated by the system administrator to the MPU 301 through the CPU 1, and the MPU 301 itself may set the master virtual disk device and the slave virtual disk device on the same real disk device. You may decide by solving the combination problem which asks for as many pairs as possible within the range which satisfies the condition which does not exist.

(3) RAID3 In the case of RAID3, the MPU 301 uses the parity disk as the virtual disk device in the real disk device having the smallest number of created virtual disk devices, and the virtual disk device created on another real disk device. Is used for each data disk to construct a disk array corresponding to RAID3. Therefore, in the present example, as shown in FIG.
The virtual disk device D31 on the real disk device D3 is a parity disk, and the virtual disk devices D11 to D14 on the real disk devices D1 and D2 other than the real disk device D3.
A disk array in which D21 and D22 are used as data disks is constructed. Of course, the system administrator may specify the virtual disk device used as the parity disk and the virtual disk device used as the data disk to the MPU 301 through the CPU 1. Then, the virtual disk devices D11 to D14, D21, D
Logical addresses are assigned to the disk areas 22 and D31 as shown in A-1 to n-6 of FIG. 11, and as disk array configuration information, for example, as shown in FIG. 12, for each logical address. , The data virtual disk device number and its disk address (offset address from the beginning), and the parity virtual disk device number and its disk address (offset from the beginning) are recorded.

Next, the read / write operation of the disk array constructed as described above will be described.

(1) RAID0 When a read request or a write request specifying a logical address is given from the CPU 1 to the control unit 3 through the interface 2, the MPU 301 of the control unit 3 causes the RAM 303 to operate.
The disk array configuration information as shown in FIG. 8 stored in FIG. 8 is searched for by its logical address, and the virtual disk device number and the in-disk address are acquired. Next, in order to find out which physical address of which real disk device the start address of the virtual disk device having the obtained virtual disk device number corresponds to, the obtained virtual disk device number is translated through the interface 306. Send to Part 4.

The address conversion unit 4 searches the disk configuration table as shown in FIG. 5 with the received virtual disk device number to obtain the real disk device number and serial number,
From the acquired serial number and the size of the virtual disk device stored in the size register 5, (serial number-1) × size is calculated, and the calculation result and the acquired real disk device number are calculated by the MPU 301 through the interface 306.
To return.

The MPU 301 obtains the address on the real disk device having the returned real disk device number by adding the in-disk address to the returned calculation result, and accesses to the address of the real disk device. Perform processing. First, it is checked whether or not there is a copy of the data of the address of the real disk device in the cache 302. If there is a read request, the data read from the cache 302 is returned to the CPU 1 through the interface 2, CP for write request
Caches the data passed from U1 at the time of write request 3
At the same time as writing to 02, it is also written to the address of the real disk device via the selector 304. Alternatively, only the cache 302 may be updated, and the data on the real disk device may be updated when the cache 302 is flushed.

On the other hand, if there is no copy of the data of the address of the real disk device in the cache 302, in the case of a read request, the data is read from the address of the real disk device through the selector 304 and the CPU 1 And the cache 302 is cached. In the case of a write request, the data passed from the CPU 1 at the time of the write request is written to the address of the real disk device via the selector 304.

(2) RAID1 (A) Read request When a read request specifying a logical address is given from the CPU1 to the control unit 3 through the interface 2, the MPU 301 of the control unit 3 is stored in the RAM 303 and stored in the RAM 303 shown in FIG.
The disk array configuration information as shown in 0 is searched by the logical address, and the master side virtual disk device number and the disk address are acquired. Next, as in the case of RAID 0, the address conversion unit 4 finds out which physical address of which real disk device the start address of the virtual disk device having the acquired master-side virtual disk device number corresponds to. The real disk device number and the real disk device address corresponding to the side virtual disk device number and the disk address are obtained. And
The access processing to the address of the real disk device is performed in the same manner as in the case of RAID0.

(B) Write request When a write request designating a logical address is given from the CPU 1 to the control unit 3 through the interface 2, the MPU 301 of the control unit 3 is stored in the RAM 303.
The disk array configuration information as shown in 0 is searched by its logical address, and the master side virtual disk device number, the slave side virtual disk device number, and the in-disk address are acquired. Next, which physical address of which real disk device the start address of the virtual disk device having the acquired master-side virtual disk device number corresponds to, and the virtual disk device having the acquired slave-side virtual disk device number Of the physical address of which real disk device corresponds to the real disk device number of the master side virtual disk device number and the in-disk address obtained by the address conversion unit 4 as in the case of RAID0. And a real disk device address, and a real disk device number and a real disk device address corresponding to the slave virtual disk device number and the disk address. Then, two access processes to the address of the real disk device are performed in the same manner as in the case of RAID0. The master virtual disk device and the slave virtual disk device are created on different real disk devices.
Updates to slaves can be done in parallel.

(3) RAID3 (A) Read request A read request that specifies, for example, A-1 as a logical address from the CPU 1 is sent through the interface 2 to the control unit 3
The MPU 301 of the control unit 3, the RAM 3
The data corresponding to each of the logical addresses A-1 to A-6 is searched by searching the disk array configuration information as shown in FIG. 12 stored in 03 with the logical address (A-x) (x is an arbitrary value). Virtual disk device number and disk address,
Also, the virtual disk device number for parity and the in-disk address are acquired.

Next, it is determined by RAID0 which physical address of which real disk device the head address of the virtual disk device having the acquired data virtual disk device number and parity virtual disk device number corresponds to.
In the same manner as in the case of 1, the real disk device number and the real disk device address corresponding to the data virtual disk device number and the disk internal address, the parity virtual disk device number and the disk internal address are obtained. The real disk device number and the real disk device address corresponding to are obtained.

Then, the read process for the address of the real disk device is performed in the same manner as in the case of RAID0 to obtain the data stored in the logical addresses A-1 to A-6 and the parity thereof. At this time, access to virtual disk devices on different real disk devices can be performed in parallel. Next, a parity check is performed. If there is no error, the data corresponding to the read-requested logical address A-1 is returned to the CPU 1, and if there is an error, that fact is returned to the CPU 1.

(B) Write request A write request specifying, for example, A-1 as a logical address from the CPU 1 is sent through the interface 2 to the control unit 3
The MPU 301 of the control unit 3, the RAM 3
The disk array configuration information as shown in FIG. 12 stored in 03 is searched by the logical address (A-1), and the virtual disk device number for data corresponding to the logical address A-1 and the address in the disk, and the parity. For virtual disk device number and disk address

Next, RAID 0 is used to determine which physical address of which real disk device the head address of the virtual disk device having the acquired data virtual disk device number and parity virtual disk device number corresponds to.
In the same manner as in the case of 1, the real disk device number and the real disk device address corresponding to the data virtual disk device number and the disk internal address, the parity virtual disk device number and the disk internal address are obtained. The real disk device number and the real disk device address corresponding to are obtained.

Then, the read process for the address of the real disk device is performed in the same manner as in the case of RAID0 to obtain the data stored at the logical address A-1 and its parity. Since the data virtual disk device and the parity virtual disk device are created on different real disk devices, they can be processed in parallel.

Next, the acquired data of the logical address A-1 and the data passed from the CPU 1 at the time of the write request,
A new parity is generated from the acquired parity, and this parity and the data passed from the CPU 1 are RA
Similar to the case of ID0, it is written back to the original location of the cache 302 or the real disk device. At this time, it is possible to process in parallel.

FIG. 13 is a block diagram of another embodiment of the disk array device of the present invention. This embodiment differs from the embodiment shown in FIG. 2 in that the virtual disk construction unit 11
A size setting register 20 that allows a system administrator to set an arbitrary value through the CPU 1, the interface 2, and the control unit 3 is provided therein, and the size calculation unit 6 has a maximum commitment of the capacities of a plurality of real disk devices used for the disk array. If the number is smaller than the value set in the size setting register 20, the value set in the size setting register 20 is determined as the size of the virtual disk device.
The obtained greatest common divisor is determined as the size of the virtual disk device.

For example, a 2 GB real disk device and 1.2
When there are two real disk devices, the real disk device of GB and the real disk device of GB, in the embodiment of FIG.
According to the MB, a total of eight virtual disk devices are generated. If, for example, 1 GB is set in the size setting register 20, the size of the virtual disk device will be 1 GB,
Three 1 GB virtual disks are created. This allows
Although 0.2 GB is unused, it is possible to avoid a situation where the size of the virtual disk device becomes extremely small.

FIG. 14 is a block diagram of still another embodiment of the disk array device of the present invention. The difference of this embodiment from the embodiment shown in FIG. 2 is that the size calculator 6 is removed and the system administrator can directly set the size of the virtual disk device in the size register 5 through the CPU 1, the interface 2 and the controller 3. There is a point in doing so. The system administrator must determine the size of the virtual disk device, which increases the workload of the system administrator, but has the advantage that the system administrator can adjust the size of the virtual disk device and the total number of virtual disk devices. is there.

[0047]

As described above, according to the present invention,
Because a disk array is configured using multiple virtual disk devices of the same capacity, including multiple virtual disk devices created on the same physical disk device, multiple physical disk devices of different capacities must be used for all disk devices. It is possible to construct a disk array corresponding to a large capacity RAID3 that is equal to or close to the total capacity.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of an embodiment of the present invention.

FIG. 2 is a block diagram of an embodiment of a disk array device of the present invention.

FIG. 3 is a block diagram of an embodiment of a control unit.

FIG. 4 is a flowchart showing an example of a procedure for constructing a disk array using a plurality of real disk devices.

FIG. 5 is a diagram showing an example of a disk configuration table.

FIG. 6 is a diagram showing a correspondence relationship between virtual disks and real disks.

FIG. 7: R configured by using a plurality of virtual disk devices
It is a figure which shows the structural example of the disk array of AID0.

FIG. 8 is an R configured by using a plurality of virtual disk devices.
It is a figure which shows the example of the disk array configuration information for the disk array of AID0.

FIG. 9 is an R configured by using a plurality of virtual disk devices.
It is a figure which shows the structural example of the disk array of AID1.

FIG. 10 is a diagram showing an example of disk array configuration information for a RAID 1 disk array configured by using a plurality of virtual disk devices.

FIG. 11 is a diagram showing a configuration example of a RAID 3 disk array configured by using a plurality of virtual disk devices.

FIG. 12 is a diagram showing an example of disk array configuration information for a RAID 3 disk array configured by using a plurality of virtual disk devices.

FIG. 13 is a block diagram of another embodiment of the disk array device of the present invention.

FIG. 14 is a block diagram of a disk array device according to still another embodiment of the present invention.

[Explanation of symbols]

1 ... CPU 2 ... Interface 3 ... Control unit 4 ... Address converter 5 ... Size register 6 ... Size calculator 7 ... Disk configuration table 10 ... Disk array controller (DAC) 11 ... Virtual disk configuration unit 20 ... Size setting register

Claims (5)

(57) [Claims] 1. When a plurality of real disk devices in which at least any two real disk devices have different capacities are divided into disk areas of the same capacity, each disk area is defined as a virtual disk device, and the same real disk device is used. Including the multiple virtual disk devices created in, one of the multiple virtual disk devices of the same capacity is used for the parity disk, and
Using all virtual disk devices of
A disk array device which constitutes one disk array corresponding to RAID 3 having a capacity which is equal to or close to the total capacity of the plurality of real disk devices. 2. A disk array device comprising a plurality of real disk devices in which at least any two real disk devices have different capacities, and a controller connected to the plurality of real disk devices, The device comprises a size register for holding a size for dividing a disk area of the plurality of real disk devices, and an individual disk area obtained by dividing the plurality of real disk devices by the size as a virtual disk device, and each virtual disk device. One of the plurality of virtual disk devices including the means for generating a disk configuration table showing the correspondence with the real disk device and the plurality of virtual disk devices generated on the same real disk device is parity-checked.
All other virtual disk devices are used for data
Using the use the disc, equal to the sum of the capacities of the plurality of physical disk devices, it means that constitutes one disk array corresponding to RAID3 substantially near capacity, read and write requests to the disk array from the host apparatus And a means for converting an address for the virtual disk device into an address for the corresponding real disk device by referring to the disk configuration table. 3. An RAI in which a virtual disk device in a real disk device having the smallest number of created virtual disk devices is used as a parity disk, and a virtual disk device created on another real disk device is used as a data disk, respectively.
The disk array device according to claim 2, wherein a disk array corresponding to D3 is configured. 4. The disk array device according to claim 2, further comprising means for obtaining the greatest common divisor of the capacities of the plurality of real disk devices, and the value obtained by the means is set in the size register. 5. A size setting register for holding a minimum size, and means for obtaining the greatest common divisor of the capacities of the plurality of real disk devices, wherein the value obtained by the means is smaller than the minimum size. 4. The disk array device according to claim 2, wherein the minimum size is set in the size register, and otherwise the obtained value is set in the size register.