JP2002157093A - Disk array device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce a resource for a disk prepared as a hot spare. SOLUTION: A plurality of physical disks are managed as one logic disk, or one portion of an area in the one physical disk is managed as one logic disk, to realize a RAID function using the logic disk. Data restoration processing is allowed to be executed by this constitution when total disk capacity prepared as the hot spare is larger than trouble-generated disk capacity, even when the hot spare of capacity same to the trouble-generated disk capacity or more is not prepared. When data restorations are required at the same time by the plural RAIDs, the data restoration processing is allowed to be executed when total restored data capacity is smaller than the disk capacity prepared as the hot spare, even when only one hot spare is prepared.

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 having a RAID configuration, and more particularly, to a disk array device capable of reducing resources of a disk prepared as a hot spare.

[0002]

2. Description of the Related Art In a disk array device adopting a RAID configuration, a disk defined as a hot spare is prepared, and when a failure occurs in a used disk, a hot spare disk is replaced with a RAID in place of the failed disk. Connect and restore the data on the failed disk from the data on the non-failed disk,
By storing the restored data on the connected disk, data restoration processing is executed.

In order to realize this data restoration processing, a conventional disk array apparatus employs a configuration in which a disk having a larger capacity than a disk in use is prepared as a hot spare disk.

That is, when a failure occurs in a used disk, a disk having a larger capacity is prepared as a hot spare disk so that the disk can be replaced with the disk.

When a plurality of RAIDs are mounted, a larger number of hot spare disks than the number of mounted RAIDs are prepared.

That is, when a failure occurs simultaneously in disks of a plurality of RAIDs to be mounted, the mounted RAIDs can be replaced with those disks.
A number of disks larger than the number of IDs are prepared as hot spare disks.

[0007]

However, according to such a conventional technique, when a part of a disk constituting the disk array device is replaced with a new one, the capacity of the new disk is changed. If the capacity of the hot spare disk is larger than the capacity of the hot spare disk, there is a problem that the hot spare disk must also be replaced in preparation for the occurrence of a failure in the new disk.

Further, according to such a conventional technique, when a plurality of RAIDs are implemented, their RAID
Considering that failures may occur at the same time in the ID, RAID in which a hot spare disk is mounted
There is a problem that a number larger than the number must be prepared.

As described above, according to the prior art, there is a problem that a great limitation is imposed on the number and capacity of disks prepared as hot spares.
There is a problem that resources of a disk prepared as a hot spare increase.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a new disk array device which can reduce the resources of a disk prepared as a hot spare.

[0011]

In order to achieve this object, a disk array device according to the present invention employs a first configuration for storing information on an unused area of each physical disk when adopting a RAID configuration. Storage means, second storage means for storing area information of physical disks constituting each logical disk, and third storage means for storing attribute information of each RAID and storing identification information of the logical disks constituting the same. In the event that a failure occurs in the storage means and the physical disk, an unused area of the physical disk in place of the failed physical disk is acquired according to the data stored in the first storage means, and is set as a new logical disk In addition to the logical disk having the failed physical disk registered in the third storage means as a constituent disk, the logical disk is registered in the second storage means. Configured to include an execution unit for registering a logical disk newly set.

According to this configuration, according to the present invention, a plurality of physical disks can be treated as one logical disk,
A part of the area of one physical disk is treated as one logical disk, and RA is used by using those logical disks.
The ID function can be realized.

That is, according to the present invention, a RAID function can be realized by treating a logical disk composed of a plurality of physical disks as one hot spare and assigning it to a failed disk. ,
For example, as shown in FIG. 1, even if a hot spare having the same capacity or more as that of the failed disk is not prepared, if the total capacity of the disks prepared as the hot spare is larger than the capacity of the failed disk, data restoration is performed. Processing can be performed.

According to the present invention, the RAID function is realized by treating a logical disk constituted by a part of the area of one physical disk as one hot spare and allocating it to a failed disk. For example, as shown in FIG.
When data restoration is required simultaneously in D, if the total amount of data to be restored is smaller than the capacity of a disk prepared as a hot spare, data restoration processing can be executed even when only one hot spare is prepared. Become like

As described above, according to the present invention, data restoration processing can be executed in a form that greatly reduces the number of disks and the capacity imposed on the capacity defined as a hot spare. The resources of the defined disk can be reduced.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail according to embodiments.

FIG. 3 shows a RAID device 1 equipped with the present invention.
1 illustrates an example of an embodiment.

As shown in FIG. 1, a RAID device 1 according to the present invention includes a physical disk 10 that is configured as a RAID or prepared as a hot spare, a basic I / O control unit 11 that performs interface processing with a higher-level device, and Physical disk 1
0, a process (Rebuild) of restoring the data of the failed physical disk 10 to the hot spare physical disk 10 or replacing the failed physical disk 10 from the hot spare physical disk 10 to the replacement disk A data restoration control unit 12 for performing a data write-back process (Copyback), a spare area management unit 13 for performing a hot spare acquisition / release process, a device control unit 14 for controlling access to a device, and a physical disk management table 15 Logical disk management table 16,
And a RAID table 17.

In the thus configured RAID device 1 of the present invention, when an access request is issued from a higher-level device, the basic I / O control unit 11 requests access to the device control unit 14 and receives the request. And the device control unit 1
4 accesses the physical disk 10 and, upon detecting the occurrence of a device failure at that time, returns an error response to the basic I / O control unit 11, and in response, the basic I / O control unit 11 Requests the control unit 12 to restore data,
In response to this, the data restoration control unit 12 requests the spare area management unit 13 to acquire a hot spare, and in response, the spare area management unit 13 acquires a hot spare and sends it to the data restoration control unit 12. In response to this, the data restoration control unit 12 requests the data restoration process to the device control unit 14, and in response, the device control unit 14 performs a process of executing the data restoration process.

Next, the physical disk management table 15, the logical disk management table 16, and the RAID table 17
Will be described.

FIG. 4 shows an embodiment of the table structure of the physical disk management table 15, and FIG. 5 shows an embodiment of the table structure of the logical disk management table 16, and FIG.
An example of a table structure of the RAID table 17 is shown in FIG.

As shown in FIG. 4, the physical disk management table 15 is associated with the physical disk number of each physical disk 10 (such as P-Disk-0 in the figure), and the manufacturer name and product name of the physical disk 10. "Device name" which is information such as
"Physical capacity" which is information on the physical capacity of the physical disk 10, and "status (0: Available, 1: Broken, 2: Hot Spare,
3: Not Available) "," Free queue terminal "which is pointer information to a Free queue for queue management of an unused area of the physical disk 10, and the physical disk 1
“Access ID”, which is information required when accessing “0”, is managed.

Fr which this "Free queue terminal" points to
The ee queue, as shown in FIG.
The queue management of the area management block that manages the unused area (represented by the offset indicating the area start point and the number of blocks indicating the area size) of the physical disk 10 makes the unused area of the physical disk 10 available. Managing.

On the other hand, the logical disk management table 16
As shown in FIG. 5, in association with the logical disk number of each logical disk (such as L-Disk-0 in the figure), the logical capacity of the logical disk (the physical disk 10 constituting the logical disk has "Logical capacity" which is information on the total physical capacity) and the RAI in which the logical disk is incorporated.
It manages “RAID number” which is information of the number D, and “Used queue terminal” which is point information to the used queue which manages the area of the physical disk 10 constituting the logical disk.

Us indicated by this "used queue terminal"
As shown in the figure, the ed queue is an area management unit that manages an area (expressed by an offset indicating the area start point and the number of blocks indicating the area size) of the physical disk 10 constituting the logical disk. By managing the blocks in the queue, the area of the physical disk 10 constituting the logical disk is managed.

On the other hand, as shown in FIG. 6, the RAID table 17 is associated with the RAID number of each RAID (such as RAID-0 in the figure), and the information of the RAID level “RAID level (0: Level 0,1: Level 1, 2: Level 0 + 1, 3: Level
3,4: Level 4,5: Level 5) "and the status information of the RAID," Status (0: Normal, 1: Degenerate, 2: Flokun, 3: Rehabilit, 4: Coherhack) ", The “disk configuration number” which is information on the number of logical disks configuring the RAID, and the R
If an L-Disk number, which is the information of the logical disk number of the logical disk that constitutes the AID, and a hot spare is used instead of the logical disk indicated by the L-Disk number, the logical disk is created from the hot spare. "HS L-Disk number", which is the information of the logical disk number of the specified logical disk
And "Status (0: Available, 1: Broken, 2: Rebuild, 3: Copy), which is information on the status of the logical disks constituting the RAID.
back, 4: HS Used) ".

For example, as shown in FIG.
When a logical disk indicated by “logical 0” is allocated to the physical disk 10 indicated by “”, a failure occurs in the logical disk, and the logical disk indicated by “logical 4” is newly indicated by “physical 4”. A logical disk indicated by “logical 1” is allocated to a physical disk 10 indicated by “physical 1”, and a logical disk indicated by “logical 2” is allocated to a physical disk 10 indicated by “physical 2”. The logical disk indicated by
When a logical disk indicated by “Logic 3” is allocated to a physical disk 10 indicated by “Physical 3”, a failure occurs in the logical disk and the logical disk indicated by “Logic 5” is newly assigned to “Physical 3”. Physical disk 1 indicated by "4"
It is assumed that the allocation is obtained from 0.

In this case, the RAID table 17
The data as shown in FIG. 8 is managed, the physical disk management table 15 manages the data as shown in FIG. 9, and the logical disk management table 16 manages the data as shown in FIG.

Here, it is assumed that the capacity of each of the physical disks 10 of “physical 0” to “physical 3” is “10”, and that the capacity of the physical disk 10 of “physical 4” is “25”.

By providing the RAID table 17 / logical disk management table 16 for managing such data, the device control unit 14 can realize the RAID function using a logical disk.

That is, the device controller 14 sets the RAI
When processing to realize the D function, first, R
The logical disk to be processed is specified by referring to the AID table 17. Subsequently, by referring to the logical disk management table 16 to specify which area of which physical disk 10 the specified logical disk is, the physical disk 1 to be actually processed is specified.
The area of 0 is specified. Subsequently, a process for implementing the RAID function is performed on the specified area of the physical disk 10.

As described above, the device control unit 14
The RAID function can be realized using a logical disk.

If a configuration is adopted in which the RAID function is realized using logical disks, as can be seen from the table structure of the logical disk management table 16 shown in FIG.
A part of the area of one physical disk 10 is treated as a logical disk, a non-contiguous area of one physical disk 10 is treated as a logical disk, or a group of areas of a plurality of physical disks 10 is a logical disk. Or a combination of a part and a part of the area of a plurality of physical disks 10 can be handled as a logical disk. Therefore, the number and capacity imposed on the physical disks 10 prepared as hot spares are limited. It is possible to greatly reduce.

Next, the processing executed by the spare area management unit 13 will be described with reference to the processing flow of FIG.

When a request to acquire a hot spare is issued from the data restoration control unit 12 by designating the capacity of the physical disk 10 in which a failure has occurred, the spare area management unit 13 firstly issues a request as shown in the processing flow of FIG. In step 1, by referring to the physical disk management table 15, the physical disk 10 having the status of hot spare and having the pointer information to the Free queue is searched to find the same as the capacity specified in the acquisition request. Find unused area of hot spare with size.

Subsequently, in step 2, it is determined whether or not the unused area of the hot spare having the same size as the capacity designated by the acquisition request has been obtained by this search processing, and it is determined that the unused area cannot be obtained. Sometimes Step 3
To find an unused area of a hot spare having a size larger than the capacity specified in the acquisition request.

Subsequently, in step 4, it is determined whether or not the unused area of the hot spare having a size larger than the capacity designated by the acquisition request can be obtained by the search processing, and it is determined that the unused area cannot be obtained. In some cases, the process proceeds to step 5 to search for a combination of unused hot spare areas having a size smaller than the capacity specified in the acquisition request and satisfying the required size.

Subsequently, at step 6, it is determined whether or not a combination of unused areas of the hot spare having a size smaller than the capacity designated by the acquisition request and satisfying the required size has been obtained by this search processing. When it is determined that the hot spare cannot be obtained, the process proceeds to step 7, where a failure to obtain the hot spare is returned to the data restoration control unit 12, and the process ends.

On the other hand, when it is determined in step 2 / step 4 / step 6 that the unused area of the hot spare has been acquired, the process proceeds to step 8 where the required size is cut out from the acquired unused area (the acquired unused area). If the used area is the same size as the requested size, it is used as it is), it is set as a new logical disk, and the management data of the physical disk management table 15 / logical disk management table 16 / RAID table 17 is updated accordingly. After that, completion of acquisition of the hot spare (acquisition success) is returned to the data restoration control unit 12, and the process is terminated.

Referring to FIGS. 12 to 17, the updating process of this table will be specifically described.

As shown in FIG. 12, a logical disk indicated by "logical 0" is allocated to a physical disk 10 indicated by "physical 0", and a "logical 1" is assigned to the physical disk 10 indicated by "physical 1". The logical disk indicated by “logical 2” is allocated to the physical disk 10 indicated by “physical 2”, and the logical disk indicated by “logical 2” is allocated to the physical disk 10 indicated by “physical 3” It is assumed that a failure has occurred in the physical disk 10 indicated by "physical 0" when a logical disk is allocated and the physical disk 10 indicated by "physical 4" or "physical 5" is prepared as a hot spare. .

When this failure occurs, in the example shown in FIG. 12A, the spare area management unit 13 replaces a part of the unused area of one physical disk 10 prepared as a hot spare with a new logical disk. It is assumed that a process of setting the disk as a logical disk ("logical 4") and replacing it with a logical disk indicated by "logical 0" is performed.

On the other hand, the spare area management unit 13
In the example shown in (b), 2 is prepared as a hot spare.
It is assumed that a process in which unused areas of two physical disks 10 are merged and set as a new logical disk (“logical 4”) and replaced with a logical disk indicated by “logical 0” is performed. ing.

When performing the replacement process shown in FIG. 12A, the spare area management unit 13 updates the physical disk management table 15 as shown in FIG. 13, and updates the logical disk management table 16 as shown in FIG. 17 and the RAID table 17 is updated as shown in FIG. Here, FIG. 13 and FIG.
In the example of No. 4, the capacity of each physical disk 10 of “physical 0” to “physical 3” is assumed to be “10”, and the capacity of the physical disk 10 of “physical 4” is assumed to be “25”.

On the other hand, the spare area management unit 13
When performing the replacement process shown in (b), the physical disk management table 15 is updated as shown in FIG.
The logical disk management table 16 is updated as shown in FIG. 16, and the RAID table 17 is updated as shown in FIG.
The update process is performed as shown in FIG. Here, FIG.
In the example of FIG. 16, the capacity of each physical disk 10 of “physical 0” to “physical 3” is assumed to be “20”, and the capacity of each physical disk 10 of “physical 4” and “physical 5” is “15”. "Is assumed.

Thus, spare area management unit 13
When a new logical disk is set by acquiring a hot spare, the physical disk management table 15
, The capacity of the hot spare registered in the logical disk is reduced, the new logical disk is additionally registered in the logical disk management table 16, and the replacement information of the logical disk is registered in the RAID table 17.

Next, in accordance with the processing flow shown in FIGS. 18 and 19, the spare area management unit 13 performs the processing performed in step 5 of the processing flow of FIG. The process of searching for a combination of used areas that satisfies the required size) will be described in detail.

When the spare area management unit 13 enters step 5 of the processing flow of FIG. 11, first, as shown in the processing flow of FIGS. 18 and 19, first, in step 1, the local area (shown in FIG. 3) ) Is initialized.

Subsequently, in step 2, the area management block having the largest number of blocks is searched from the area management blocks linked to the Free queue of the physical disk 10 whose status indicates the hot spare. When it is determined whether or not the block has been searched for, and it is determined that the block cannot be searched, that is, when it is determined that the hot spare physical disk 10 has no unused area, the unacquired acquisition ends.

On the other hand, if it is determined in step 3 that the area management block having the maximum number of blocks has been searched, the flow advances to step 4 to remove the searched area management block from the queued Free queue. hand,
Queue to the acquisition queue.

Subsequently, in step 5, the "acquisition flag (prepared as will be described later,
If an unused area exceeding the requested size cannot be acquired, “0” is displayed) ”is set to“ 0 ”, and in the subsequent step 6,“ search pointer (physical disk management table 15) is initialized (set to the head).

Subsequently, in step 7, the physical disk management table 15 is searched with the search pointer as a starting point.
A search is made for an area management block linked to the Free queue of 0, and the search pointer is updated accordingly.

Subsequently, in step 8, the number of blocks of the searched area management block and the total number of blocks of the area management block queued in the acquisition queue are added. It is determined whether the calculated addition value is larger than the required size.

When it is determined that the calculated addition value is smaller than the required size, the process proceeds to step 17 to determine whether the search pointer indicates that the search has been completed to the end. If it is determined that the search has not been completed to the end, the process returns to step S7.

As described above, in step 2, when the area having the largest number of blocks is searched from the unused areas of the hot spare (those not queued in the acquisition queue), one area management block is found in the acquisition queue. In the example of the processing stage in which (unused area) is queued, while the search pointer is advanced, it is determined whether there is another unused area to be added to which exceeds the requested size. That is, processing is performed so as to search for or.

By continuing this search processing, if it is determined in step 9 that an unused area in which the calculated added value is larger than the required size has been searched, the process proceeds to step 10 where the acquisition flag is set. When it is determined whether or not "0" is displayed, and when it is determined that "0" is displayed, the process proceeds to step 11, where the area management block searched in step 7 is queued. Free
The queue is removed from the queue and queued at the end of the acquisition queue. In the following step 12, the acquisition flag is set to "1".

That is, in the example of the processing stage in which one area management block (unused area) is queued in the acquisition queue, the unused area searched in step 2 according to the processing of step 9 / step 10 When it is determined for the first time that another unused area to be added to the area that can exceed the required size can be searched, the searched area management block (unused area) is added to the queue at the end of the acquisition queue. At the same time, the acquisition flag is set to "1" to indicate that.

Then, the process proceeds to a step 17, wherein it is determined whether or not the search pointer indicates that the search has been completed to the end, and it is determined that the search has not been completed to the end. Sometimes, the process returns to step 7.

On the other hand, when it is determined in step 10 that the acquisition flag is displaying "1" instead of "0", that is, it is determined that a search for achieving a size exceeding the required size can be further performed. When making a decision, step 1
Proceeding to 3, the number of blocks in the area management block retrieved in step 7 is compared with the number of blocks in the last area management block in the acquisition queue, and the following step 14 is performed.
Then, it is determined whether or not the number of blocks in the last area management block of the acquisition queue is larger.

When it is determined that the number of blocks in the last area management block of the acquisition queue is larger by this determination processing, the process proceeds to step 15, where the last area management block is removed from the acquisition queue. Is queued in the physical disk management table 15 (return to the original state).

Subsequently, in step 16, the area management block searched in step 7 is queued in the Free queue.
Dequeue from the queue and queue at the end of the acquisition queue. In the following step 17, it is determined whether the search pointer indicates that the search has been completed, and that the search has not been completed. When it is determined that the above is indicated, the process returns to step S7.

On the other hand, when it is determined in step 14 that the number of blocks in the last area management block of the acquisition queue is smaller, the process immediately proceeds to step 17 without performing the processing in step 15 / step 16. If it is determined whether the search pointer indicates that the search has been completed to the end, and if it indicates that the search has not been completed to the end, the process returns to step S7.

That is, in the example of the processing stage in which one area management block (unused area) is queued in the acquisition queue, the unused area searched in step 2 according to the processing in step 9 / step 10 When judging that another unused area to be added to the area can be searched for that achieves the required size, the unused area having the smaller size should be selected as much as possible. Is processed.

On the other hand, when it is determined in step 17 that the search pointer indicates that the search has been completed to the end, the process proceeds to step 18 to determine whether or not the acquisition flag indicates "1". When it is determined that “0” is being displayed, that is, when it is determined that the process of step 18 has been reached without performing the processes of steps 11 to 16, one In the example of the processing stage in which the area management block (unused area) is queued, it is determined that the combination of two unused areas cannot exceed the required size, and three unused areas are determined. The processing returns to step 2 in order to search whether there is a combination that can exceed the required size.

As described above, when returning to step 2, in the example of the processing stage in which one area management block (unused area) is queued in the acquisition queue, in step 2, The processing is performed so as to search the area management block having the maximum number of blocks from the area management blocks of the hot spare not queued in the queue.

On the other hand, if it is determined in step 18 that the acquisition flag indicates "1", the unused area of the requested size is acquired (the area of the requested size is deleted from the one queued in the acquisition queue). It is determined that the extraction has succeeded, and the process ends.

Thus, spare area management unit 13
When searching for a combination of unused hot spare areas with a size smaller than the capacity specified in the acquisition request that satisfies the required size, it is requested that the number of unused areas to be acquired be as small as possible. In addition to selecting the used area, processing is performed to select the unused area so that the excess size of the obtained unused area is as small as possible, and to execute the search processing.

By using this search method, the number of area management blocks queued in the logical disk management table 16 is reduced, thereby reducing the load of the search processing required for realizing the RAID function. The speed can be increased, and in some cases, an unused area corresponding to the required size can be acquired, thereby making it possible to eliminate unnecessary use of the unused area.

[0069]

As described above, according to the present invention,
Data restoration processing can be executed in a form that greatly reduces the number and capacity of disks defined as hot spares, so that the resources of disks defined as hot spares can be reduced. Become like

For example, by treating a logical disk composed of a plurality of physical disks as one hot spare and assigning it to a failed disk, RAID
Since the function can be realized, even if a hot spare with the same capacity or more as the failed disk is not prepared,
If the total capacity of the disks prepared as hot spares is larger than the capacity of the failed disk, data restoration processing can be executed.

For example, a RAID function can be realized by treating a logical disk constituted by a part of the area of one physical disk as one hot spare and assigning it to a failed disk. Therefore, if it is necessary to simultaneously restore data in a plurality of RAIDs, and if the total amount of data to be restored is smaller than the capacity of a disk prepared as a hot spare, the data is restored even when only one hot spare is prepared. The restoration process can be executed.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of the present invention.

FIG. 2 is an explanatory diagram of the present invention.

FIG. 3 is an embodiment of the present invention.

FIG. 4 is an embodiment of a table structure of a physical disk management table.

FIG. 5 is an example of a table structure of a logical disk management table according to an embodiment;

FIG. 6 is an example of a table structure of a RAID table according to an embodiment;

FIG. 7 is an example of a RAID configuration.

FIG. 8 is an example of management data of a RAID table.

FIG. 9 is an example of management data of a physical disk management table.

FIG. 10 is an example of management data of a logical disk management table.

FIG. 11 is an embodiment of a processing flow executed by a spare area management unit;

FIG. 12 is an explanatory diagram of a table update process.

FIG. 13 is an explanatory diagram of a table update process.

FIG. 14 is an explanatory diagram of a table update process.

FIG. 15 is an explanatory diagram of a table update process.

FIG. 16 is an explanatory diagram of a table update process.

FIG. 17 is an explanatory diagram of a table update process.

FIG. 18 is an embodiment of a processing flow executed by a spare area management unit;

FIG. 19 is an embodiment of a processing flow executed by a spare area management unit;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 RAID device 10 Physical disk 11 Basic I / O control unit 12 Data restoration control unit 13 Spare area management unit 14 Device control unit 15 Physical disk management table 16 Logical disk management table 17 RAID table

Claims (5)

[Claims] In a disk array device adopting a RAID configuration, a first storage device stores information on an unused area of each physical disk.
Storage means for storing area information of physical disks constituting each logical disk, and third storage means for storing attribute information of each RAID and storing identification information of the logical disks constituting the same. When a failure occurs in a physical disk, an unused area of a physical disk in place of the failed physical disk is acquired in accordance with the storage data of the first storage means, and a new logical disk And registers the newly set logical disk in place of the logical disk having the failed physical disk registered in the third storage unit as a constituent disk. A disk array device comprising: an execution unit for performing registration. 2. The disk array device according to claim 1, wherein the execution unit acquires a part of the area of the physical disk prepared for the spare, thereby replacing the physical disk with the failed physical disk. A disk array device characterized by processing to acquire an unused area of a disk. 3. The disk array device according to claim 1, wherein the execution means acquires a plurality of areas of one or more physical disks prepared for the spare, thereby obtaining a physical area in which a failure has occurred. A disk array device characterized by processing to acquire an unused area of a physical disk instead of a disk. 4. The disk array device according to claim 3, wherein said execution means selects an area so that the number of areas to be acquired is as small as possible, and replaces the physical disk with a failed physical disk. A disk array device characterized by performing processing so as to acquire an unused area of the disk array. 5. The disk array device according to claim 3, wherein the execution means selects an area so that the number of areas to be acquired is as small as possible, and enables an excess size of the area to be acquired. A disk array device characterized by performing processing so as to obtain an unused area of a physical disk in place of a failed physical disk while selecting an area to be as small as possible.