JP2010244130A - Disk array device and disk array control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To maintain responsiveness to host access even when any obstacle occurs in a portion of disk devices. <P>SOLUTION: A disk array device 10 includes: a replication management unit 250 for generating a backup logical disk from an RAID array different from an RAID array configuring a master logical disk in response to a replication instruction; and an RAID control unit 260 for, when the master logical disk and the backup logical disk are put in a synchronized state to host access, and when it is detected that any obstacle occurs in the disk device configuring the master logical disk, substituting the disk device in which the obstacle has occurred in the master logical disk with the disk device in which any obstacle has not occurred in the backup logical disk, and for changing the synchronized state of the master logical disk and the backup logical disk to a separated state to the host access. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a disk array device and a disk array control method that can maintain response performance to host access even if a failure occurs in some of the disk devices.

  A disk array device connects multiple disk devices such as hard disk drives in parallel, and controls them as a single storage device, thereby speeding up data reading and writing and increasing durability against failures. . For example, by using a technique such as RAID (Redundant Array of Inexpensive Disks), even if a failure occurs in some of the disk devices, the restoration process can be executed.

  Incidentally, the disk array device is an external storage device that reads and writes data according to instructions from the host device. Therefore, if there is a host access during the restoration process of some disk devices, the head operates randomly, and the restoration process is delayed. Also, the host access response is delayed.

JP 2002-259062 A

  However, it is necessary to control the disk array device so that performance degradation does not always occur.

  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 and a disk array control method capable of maintaining response performance to host access even if a failure occurs in some disk devices. To do.

  The present invention takes the following means in order to solve the above problems.

  A first invention is a disk array device comprising a set of disk devices in which a master logical disk and a backup logical disk managed in units of slices are different, and shows slices of the master logical disk and the backup logical disk Address storage means for storing address management information in which a logical address and a physical address of a disk device are associated with each other, failure detection means for detecting whether or not a failure has occurred in a disk device constituting the master logical disk, and the master When the logical disk and the backup logical disk are in synchronization with the host access and the failure detection means detects that a failure has occurred in the disk device constituting the master logical disk, the master logical disk And the backup logical disk By replacing the slices, a replacement unit that replaces the disk device in which the failure of the master logical disk has failed and a disk device in which the failure of the backup logical disk has not occurred; and when the replacement process is executed, the master logical disk And a state changing means for changing the backup logical disk to a disconnected state with respect to host access.

  According to a second aspect of the invention, address storage means for storing address management information in which a logical address corresponding to a slice of a master logical disk and a backup logical disk and a physical address of a disk device are associated, and a master logical disk Means for generating a backup logical disk from a disk and a set of disk devices different from the set of disk devices constituting the master logical disk, and writing address management information of each generated logical disk to the address storage means; In response to an instruction, means for copying data of the master logical disk to the backup logical disk, failure detection means for detecting whether a failure has occurred in a disk device constituting the master logical disk, and the master logical disk And the backup logical data If the failure detection means detects that a failure has occurred in the disk device constituting the master logical disk when the disk is in a synchronized state with respect to host access, the master logical disk and the backup logical disk And replacing the disk device in which the failure of the master logical disk has failed and the disk device in which the failure of the backup logical disk has not occurred, and when the replacement process is executed, A disk array device comprising: a disk and a state changing unit that changes the backup logical disk to a disconnected state with respect to host access.

  According to a third aspect of the invention, there is provided address storage means for storing address management information in which a logical address corresponding to a slice of a master logical disk and a backup logical disk and a physical address of a disk device are associated, and a master logical disk A disk and a backup logical disk are generated from a set of disk devices, and address management information of each generated logical disk is written to the address storage means, and data of the master logical disk is transferred to the backup logical disk in response to a copy command When the same data as the slice data of the master logical disk is stored in the slice of the backup logical disk, the slice of the backup logical disk is discarded and the disk device constituting the master logical disk is copied. Means for setting a new slice of the backup logical disk in a different disk device, copying the data of the master logical disk to the backup logical disk in which the new slice is set, and a disk device constituting the master logical disk A failure detection means for detecting whether or not a failure has occurred, and a disk constituting the master logical disk by the failure detection means when the master logical disk and the backup logical disk are in a synchronized state with respect to host access. When it is detected that a failure has occurred in the device, the failure of the master logical disk and the backup logical disk have occurred by replacing the slices of the master logical disk and the backup logical disk. Not disk A disk array device comprising: replacement means for replacing a device; and state changing means for changing the master logical disk and the backup logical disk into a disconnected state with respect to host access when the replacement process is executed. .

  According to the present invention, response performance to host access can be maintained even if a failure occurs in some disk devices. Further, the data restoration operation to the spare disk device due to the disk device failure and the data restoration operation by replacing the failed disk device can be executed without being affected by the host access.

1 is a schematic diagram showing a configuration of a disk array device 10 according to a first embodiment of the present invention. It is a schematic diagram for demonstrating the function implement | achieved by CPU21 which concerns on the embodiment. It is a schematic diagram which shows an example of the data structure which the address management part 240 which concerns on the embodiment hold | maintains. It is a schematic diagram for demonstrating the concept of a slice. 4 is a flowchart for explaining a replication method of the disk array device 10 according to the embodiment. 4 is a flowchart for explaining an operation when a failure occurs in the disk array device 10 according to the embodiment. 4 is a flowchart for explaining an operation when a failure occurs in the disk array device 10 according to the embodiment. It is a flowchart for demonstrating the replication method which concerns on the 2nd this embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<First Embodiment>
FIG. 1 is a schematic diagram showing the configuration of a disk array device 10 according to the first embodiment of the present invention. The disk array device 10 includes an array controller 20 and a pool 30 composed of a plurality of RAID arrays. The disk array device 10 is connected to the host device 5 via the connection path 1 and stores data in response to a request from the host device 5.

  The array controller 20 includes a CPU 21, a program memory 22, a cache memory 23, a host connection unit 24, and an HDD connection unit 25. The CPU 21 is an arithmetic device that executes various processes.

  The program memory 22 is a storage medium that stores a storage area for programs and data. When the program stored in the program memory 22 is read by the CPU 21, each function of the array controller 20 described later is realized.

  The cache memory 23 functions as a buffer for an input / output request to the RAID array from the computer or other data storage device via the connection path 1. For example, in the write processing from the host device 5 to the RAID array, the write data is transferred to the RAID array after the data is entered in the cache memory 23. Also for the read process, if data is entered in the cache memory 23, the data is transmitted to the host device 5. If no data is entered in the cache memory 23, the data is read from the HDD 30 to the cache memory 23.

  The host connection unit 24 is an interface that connects the host device 5 and the array controller 20.

  The HDD connection unit 25 is an interface that connects the HDDs constituting each RAID array and the array controller 20.

  The pool 30 is a set of RAID arrays 310 to 360 configured by a plurality of disk devices which are so-called hard disk drives (HDDs). Here, the logical disk is divided into defined slice units. Note that slices are reserved for the entire capacity when a logical disk is created.

  In the disk array device 10 according to the present embodiment, the master logical disk and backup logical disk managed in units of slices are configured from different RAID arrays. For example, a master logical disk is used as an external storage device, and a backup logical disk is used for data recovery of the master logical disk or a backup copy source to a tape device.

  FIG. 2 is a schematic diagram for explaining functions realized by the CPU 21. When the program stored in the program memory 22 is read, the CPU 21 reads the host I / F control unit 210, the IO management unit 220, the cache memory management unit 230, the address management unit 240, the replication management unit 250, and the RAID control unit 260. And the disk I / F control unit 270.

  The host I / F control unit 210 controls the host connection unit 24 and accepts an IO request from the host device 5 and an IO request to the host device 5. The IO management unit 220 performs management of IO access from the host device 5 and IO access area control. The cache memory management unit 230 performs IO access management in the cache memory 23.

  The address management unit 240 stores “address management information” in which logical addresses corresponding to slices of the master logical disk and the backup logical disk are associated with physical addresses of the HDD. Then, the address management unit 240 uses the address management information to perform a conversion process between a logical address and a physical address. For example, when receiving an IO request from the host device 5, the address management unit 240 reads a slice corresponding to the logical address (LBA) of each logical disk, and uses the RAID array number and sequence number in the slice management data structure. Convert the data and request the RAID control unit 260 for IO processing. An example of the data structure held by the address management unit 240 is as shown in FIG. That is, the address management unit 240 stores a physical list pointer as a logical disk management data structure in association with a logical disk number. In addition, as a slice management data structure, a slice sequence number and a next sequence pointer are stored in association with the RAID array number.

  Furthermore, the address management unit 240 has a function of setting a slice S by dividing a logical disk by a predetermined size D when creating a RAID array (see FIG. 4). Then, slice sequence numbers are assigned in the order of logical addresses on the RAID array and managed by the slice management data structure. Specifically, the address management unit 240 secures slices for the capacity specified at the time of logical disk creation, and sequentially connects them to the logical disk management data structure. Here, the slice to be secured may be arbitrary, or may be selected using factors such as reliability such as a RAID level, performance such as considering the type of HDD to be configured and the inner and outer circumferences of the HDD.

  Note that the address management unit 240 enables conversion from a logical address to an address of a RAID array to which the corresponding address is assigned, but may adopt a means capable of high-speed conversion in a hash or tree format.

  The replication management unit 250 controls replication. Specifically, the replication management unit 250 generates a master logical disk and a backup logical disk from a RAID array different from the RAID array that constitutes the master logical disk in response to a replication command from the host device 5 or the like. . In addition, the replication management unit 250 writes the generated address management information of each logical disk in the address management unit 240.

  Note that “replication” means creating a backup logical disk that is a duplicate of the master logical disk. The replication state includes a “synchronization state (sync)” and a “detachment state (split)”. The synchronization state is a state in which write data from the host device 5 to the master logical disk is simultaneously written to the backup logical disk (mirroring), and the contents of both are always the same. The disconnected state is a state in which, with respect to access from the host device 5, data is written only to the master logical disk and not written to the backup logical disk.

  In addition, when receiving a copy command or the like from the host device 5, the replication management unit 250 copies the data on the master logical disk to the backup logical disk. Here, the replication management unit 250 stores a “difference management bitmap” indicating the difference between the data on the master logical disk and the data on the backup logical disk, and the master logical disk and the backup logical disk as a replication pair. to manage. In the initial state where the replication management unit 250 creates a replication, the difference between the master logical disk and the backup logical disk is 100%. Further, in the initial state where the replication is created, the master logical disk and the backup logical disk are in a “detached state”.

  If the master logical disk and the backup logical disk are in a synchronized state when the replacement processing is executed by the RAID control unit 260, the replication management unit 250 changes both of them to a disconnected state for host access. When the replication management unit 250 shifts from the synchronized state to the disconnected state, the dirty data on the cache at that time is written in the master logical disk and the backup logical disk, and the data on the backup side is determined.

  In addition, when the RAID control unit 260 performs failure recovery processing for the backup logical disk, the replication management unit 250 brings the backup logical disk and the master logical disk into a synchronized state. When the replication management unit 250 shifts from the disconnected state to the synchronous state, the replication management unit 250 refers to the difference management bitmap and copies the data of the area determined to be the difference from the master logical disk to the backup logical disk.

  The RAID control unit 260 performs control related to RAID and has a failure detection function, a slice replacement function, and a failure recovery function. The RAID control unit 260 detects whether or not a failure has occurred in the disk device constituting the master logical disk by the failure detection function.

  Further, when the RAID controller 260 detects that a failure has occurred in the disk device constituting the master logical disk when the master logical disk and the backup logical disk are in synchronization with the host access, the slice replacement is performed. By replacing the slices of the master logical disk and the backup logical disk by the function, the RAID array of the master logical disk that has been degraded due to the failure of the HDD is replaced with the RAID array of the backup logical disk.

  On the other hand, when the RAID controller 260 detects that a failure has occurred in the disk device constituting the master logical disk when the master logical disk and the backup logical disk are disconnected from the host access, the RAID controller 260 performs differential management. Based on the bitmap for use, the RAID array corresponding to the slice in which no difference is generated among the slices of the master logical disk and the backup logical disk is replaced.

Further, the RAID control unit 260 executes failure recovery processing for the backup logical disk that has been disconnected by the replication management unit 250 by the failure recovery function.
The disk I / F control unit 270 controls the HDD connection unit 25 and controls access to each of the RAID arrays 310 to 360.

Next, the operation of the disk array device 10 according to the present embodiment will be described.
In the disk array device 10, the replication management unit 250 generates a backup logical disk that is a duplicate of the master logical disk. Since the data difference is 100% in the initial replication state, the entire data is copied from the master logical disk to the backup logical disk.

FIG. 5 is a flowchart for explaining the replication method of the disk array device 10 according to the present embodiment.
First, upon receiving a replication command from the host device 5, the replication management unit 250 straddles the start address where the copy address is aligned with the logical disk with the slice size D or the boundary where the copy address + copy size is aligned with the slice size D. It is determined whether or not (S1).

  As a result of the determination, if the boundary is crossed, the copy area has reached a new slice. In this case, the replication management unit 250 reads the RAID array numbers of the master logical disk and the backup logical disk from the address management unit 240. At this time, if the RAID array numbers of the master logical disk and the backup logical disk are the same (S2-Yes), the replication management unit 250 discards the backup slice and creates a new slice from a RAID array different from the master logical disk. Secure (S3).

  Then, the replication management unit 250 newly allocates a slice secured from other than the RAID array belonging to the master logical disk to the backup logical disk (S4), and copies the data of the master logical disk (S5). At this time, the RAID address of the backup logical disk is updated (S6). This control is executed until copying of all slices of the master logical disk is completed (S7).

  As a result, when the data of the master logical disk is copied to the backup logical disk, the RAID arrays constituting the master logical disk and the backup logical disk are always different.

Next, the operation when a failure occurs in the disk array device 10 will be described.
When the RAID array is degenerated due to a failure of the HDD constituting the RAID array, restoration to the spare disk is performed. This restoration is performed by simply copying the data of the failed HDD from the remaining shadow HDD (RAID 1), reading and reproducing the remaining HDD data and parity (RAID 5), and writing to the spare HDD. It is.

FIG. 6 is a flowchart for explaining the operation when a failure occurs in the disk array device 10 according to the present embodiment. Here, a case where the replication state at the time of failure is “synchronized state” will be described.
First, when it is detected by the RAID control unit 260 that a failure has occurred in the HDD, the RAID array numbers of the slices constituting the master logical disk and the backup logical disk are read from the address management unit 240 in ascending order of logical addresses, The RAID array number of the slice in which the failure has occurred is confirmed (T1). At this time, if the master logical disk slice belongs to the degraded RAID array, the backup logical disk does not belong to the degraded RAID array. This is because the master logical disk and the backup logical disk are configured by different RAID arrays.

  In such a case (T2-Yes), the RAID control unit 260 replaces the slice corresponding to the degenerated RAID array in the master logical disk with the slice of the backup logical disk (T3).

  Then, this operation is performed on the entire area of the master logical disk (T4, T5). If even one set of slice replacement occurs (T6-Yes), the replication management unit 250 changes the master logical disk and the backup logical disk to a disconnected state (T7). As a result, no slice belonging to the RAID array being restored exists in the slice group constituting the master logical disk. Accordingly, it is possible to eliminate the mutual influence between the host I / O and the restoration process on the master logical disk.

FIG. 7 is a flowchart for explaining the operation when a failure occurs in the disk array device 10 according to the present embodiment. Here, a case will be described in which the replication state at the time of the failure is “detached state”.
Also in this case, the RAID control unit 260 reads the RAID array numbers of the slices constituting the master logical disk and the backup logical disk from the address management unit 240 in the ascending order of the logical addresses, and the RAID array numbers of the failed slices are read. Confirm (U1).

  At this time, if the slice of the master logical disk belongs to the degenerated RAID array (U2-Yes), whether or not a difference has occurred in the area on the logical disk to which the corresponding slice is assigned by the difference management bitmap. Is determined (U3).

  If no difference is found as a result of the determination (U3-Yes), the RAID control unit 260 replaces the slice corresponding to the degenerated RAID array in the master logical disk with the slice of the backup logical disk (U4). This operation is performed on the entire area of the master logical disk (U5, U6).

  As a result, it is possible to reduce the number of slices belonging to the RAID array being restored in the slice group constituting the master logical disk. Accordingly, it is possible to reduce the mutual influence between the host I / O and the restoration processing on the master logical disk.

  As described above, the disk array device 10 according to the present embodiment includes the replication management unit 250 that generates a backup logical disk from a RAID array that is different from the RAID array that configures the master logical disk, and the master in response to the replication command. When it is detected that a failure has occurred in the disk device constituting the master logical disk when the logical disk and the backup logical disk are in synchronization with the host access, the disk device in which the failure of the master logical disk has occurred A disk in which a failure has occurred because it includes a RAID controller 260 that replaces the disk device in which no failure has occurred in the backup logical disk and changes the master logical disk and the backup logical disk to a disconnected state for host access. With the exception of location from a master logical disk, it is possible to block the host access to the disk device failed, even if a part of the disk drive fails can maintain the response performance to the host access.

  In short, in the disk array device 10 according to the present embodiment, when backing up the master logical disk, the backup logical disk is configured to be mirrored from slices belonging to different RAID arrays. When the snap acquisition request is received from the host device, the dirty data in the cache is reflected on both disks. If the master logical disk and the backup logical disk are in synchronization when a certain RAID array fails, the slices belonging to the master logical disk among the slices belonging to the failed RAID array are mirrored to the backup logical disk. Replace the disk with a slice, and put replication in a disconnected state. Therefore, even if a failure occurs in the RAID array, the master logical disk does not have a degenerated RAID array slice, so that a reduction in restoration performance due to a disk device failure can be avoided. In addition, in order to put replication in a disconnected state, host access processing is performed only for the master logical disk. Therefore, it is possible to prevent a decrease in response performance to host access as compared with the synchronization state.

  In the disk array device 10, the RAID control unit 260 executes failure recovery processing for the backup logical disk that has been disconnected. When the backup logical disk is recovered from a failure, the backup logical disk and the master logical disk are synchronized. Thereby, it is possible to execute failure recovery processing for the RAID array in which degeneration occurs while the master logical disk is operating. Therefore, according to such a disk array device, the influence on the host access performance by the restoration process of the spare disk accompanying the degeneration of the RAID array and the mutual interference of the influence from the host access on the restoration performance to the spare disk are minimized. Can be limited.

<Second Embodiment>
A replication method according to the second embodiment will be described with reference to FIG. In this embodiment, it is assumed that the logical disk configuration method is virtual. The term “virtual” refers to a method of increasing capacity efficiency by not allocating all slices at the time of logical disk configuration but by allocating slices sequentially only to the portion written from the host device.

  In response to the replication command, the replication management unit 250 generates a master logical disk and a backup logical disk from the RAID array, and writes the generated address management information of each logical disk to the address management unit 240.

  Further, when the replication management unit 250 copies the master logical disk data to the backup logical disk according to the copy command, the same data as the slice data of the master logical disk is stored in the backup logical disk slice. If so, destroy the slice of the backup logical disk. Then, a new slice of the backup logical disk is set in a disk device different from the disk device constituting the master logical disk, and the master logical disk data is copied to the backup logical disk in which the new slice is set.

FIG. 8 is a flowchart for explaining the replication method according to this embodiment.
First, when a copy command is received from the host device 5, the replication management unit 250 determines whether the copy address crosses the start address where the logical disk is aligned with the slice size D or the boundary where the copy address + copy size is aligned with the slice size D. It is determined whether or not (V1).

  As a result of the determination, if the boundary is crossed, the copy area has reached a new slice. In this case, the replication management unit 250 checks whether the RAID array numbers of the master logical disk and the backup logical disk have already been assigned by referring to the address management information of the address management unit 240 (V2).

  Here, the replication management unit 250 is a case where the RAID array number of the master logical disk is not assigned (V2-No), and the RAID array number of the backup logical disk is already assigned (V3-Yes). Discards the slice of the backup logical disk and skips copying for the slice size (V4).

  If only the RAID array number of the backup logical disk is not assigned (V2-Yes, V5-No), the replication management unit 250 secures a new slice from a RAID array different from the RAID array of the master logical disk. (V8) Copy is executed (V9). Then, the correspondence relationship between the RAID array number of the backup logical disk and the logical address is updated (V10), and the above processing is repeated until the copying of all areas is completed (V11-No).

  On the other hand, if the RAID array number has already been assigned to both the master logical disk and the backup logical disk (V2-Yes, V5-Yes), the replication management unit 250 determines whether or not both RAID array numbers are the same. Is determined (V6). If it is determined that the RAID array numbers are the same (V6-Yes), the address management unit 240 discards the slice of the backup logical disk and secures a new slice from a different RAID array for the backup logical disk. (V7).

  As described above, the disk array device 10 according to the present embodiment backs up the same data as the slice data of the master logical disk when copying the data of the master logical disk to the backup logical disk according to the copy command. If it is stored in a logical disk slice, discard the backup logical disk slice, set a new slice for the backup logical disk on a disk unit different from the disk unit that configures the master logical disk, and set a new slice. Since the replication manager 250 is provided to copy the master logical disk data to the backup logical disk, the same logical address of the replication pair when the data copy from the master logical disk to the backup logical disk is completed. Vinegar RAID array can guarantee always different.

  The operation when a failure occurs in the disk array device 10 is the same as in the first embodiment. That is, when the RAID controller 260 detects that a failure has occurred in the disk device constituting the master logical disk when the master logical disk and the backup logical disk are in synchronization with the host access, the master logical disk The disk device in which the disk failure has occurred is replaced with a disk device in which the failure of the backup logical disk has not occurred, and the master logical disk and the backup logical disk are changed to the disconnected state for host access.

  Therefore, in the disk array device 10 according to the present embodiment, when copying data, even if the master logical disk and the backup logical disk are not composed of different sets of disk devices, the failed disk device is designated as the master. In addition to being removed from the logical disk, host access to the disk device in which the failure has occurred can be blocked.

<Others>
In addition, this invention is not limited to the said embodiment, In the implementation stage, it can embody by modifying a component in the range which does not deviate from the summary. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine a component suitably in different embodiment.

  The method described in the above embodiment can be stored and distributed in a storage medium such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory as a program that can be executed by a computer. In addition, as long as the storage medium can store a program and can be read by a computer, the storage format may be any form.

  Further, an operating system running on the computer based on instructions of a program installed in the computer from the storage medium, middleware such as database management software, network software, etc., part of each process for realizing the above embodiment May be executed.

  Furthermore, the storage medium in the present invention is not limited to a medium independent of a computer, but also includes a storage medium in which a program transmitted via a LAN, the Internet, or the like is downloaded and stored or temporarily stored.

  Further, the number of storage media is not limited to one, and the case where the processing in the above embodiment is executed from a plurality of media is also included in the storage media in the present invention, and the media configuration may be any configuration.

  The computer according to the present invention executes each process in the above-described embodiment based on a program stored in a storage medium, and is a single device such as a personal computer or a system in which a plurality of devices are connected to a network. Any configuration may be used.

  In addition, the computer in the present invention is not limited to a personal computer, but includes an arithmetic processing device, a microcomputer, and the like included in an information processing device, and is a generic term for devices and devices that can realize the functions of the present invention by a program. .

  5 ... Host device, 10 ... Disk array device, 20 ... Array controller, 21 ... CPU, 22 ... Program memory, 23 ... Cache memory, 24 ... Host connection unit, 25 ... HDD connection unit, 30 ... pool, 210 ... host I / F control unit, 220 ... IO management unit, 230 ... cache memory management unit, 240 ... address management unit, 250... Replication management unit, 260... RAID control unit, 270... Disk I / F control unit, 310 to 360.

Claims (5)

A disk array device composed of a set of disk devices in which the master logical disk and backup logical disk managed in units of slices are different,
Address storage means for storing address management information in which logical addresses indicating slices of the master logical disk and the backup logical disk are associated with physical addresses of disk devices;
Failure detection means for detecting whether or not a failure has occurred in the disk device constituting the master logical disk;
When the master logical disk and the backup logical disk are in a synchronized state with respect to host access, when the failure detection unit detects that a failure has occurred in the disk device constituting the master logical disk, the master logical disk A replacement means for replacing the disk device in which the failure of the master logical disk has failed and the disk device in which the failure of the backup logical disk has not occurred by replacing the slices of the logical disk and the backup logical disk;
When the replacement process is executed, state changing means for changing the master logical disk and the backup logical disk to a disconnected state for host access;
A disk array device comprising: Address storage means for storing address management information in which logical addresses corresponding to slices of a master logical disk and a backup logical disk are associated with physical addresses of disk devices;
In response to the replication command, a backup logical disk is generated from a master logical disk and a set of disk units different from the set of disk units constituting the master logical disk, and the address management information of each generated logical disk is stored in the address Means for writing to the storage means;
Means for copying data of the master logical disk to the backup logical disk in response to a copy instruction;
Failure detection means for detecting whether or not a failure has occurred in the disk device constituting the master logical disk;
When the master logical disk and the backup logical disk are in a synchronized state with respect to host access, when the failure detection unit detects that a failure has occurred in the disk device constituting the master logical disk, the master logical disk A replacement means for replacing the disk device in which the failure of the master logical disk has failed and the disk device in which the failure of the backup logical disk has not occurred by replacing the slices of the logical disk and the backup logical disk;
When the replacement process is executed, state changing means for changing the master logical disk and the backup logical disk to a disconnected state for host access;
A disk array device comprising: The disk array device according to claim 1 or 2,
Means for executing failure recovery processing on the backup logical disk that has been put into a disconnected state by the state changing means;
Means for synchronizing the backup logical disk with the master logical disk when the backup logical disk is recovered from a failure;
A disk array device comprising: Address storage means for storing address management information in which logical addresses corresponding to slices of a master logical disk and a backup logical disk are associated with physical addresses of disk devices;
In response to the replication command, a master logical disk and a backup logical disk are generated from a set of disk devices, and address management information of each generated logical disk is written to the address storage means;
When copying the master logical disk data to the backup logical disk in response to a copy command, if the same data as the slice data of the master logical disk is stored in the backup logical disk slice, the backup logical disk The slice of the logical disk is discarded, a new slice of the backup logical disk is set in a disk device different from the disk device constituting the master logical disk, and the master logical disk is set in the backup logical disk in which the new slice is set A means of copying the data of
Failure detection means for detecting whether or not a failure has occurred in the disk device constituting the master logical disk;
When the master logical disk and the backup logical disk are in a synchronized state with respect to host access, when the failure detection unit detects that a failure has occurred in the disk device constituting the master logical disk, the master logical disk A replacement means for replacing the disk device in which the failure of the master logical disk has failed and the disk device in which the failure of the backup logical disk has not occurred by replacing the slices of the logical disk and the backup logical disk;
When the replacement process is executed, state changing means for changing the master logical disk and the backup logical disk to a disconnected state for host access;
A disk array device comprising: A disk having address storage means for storing address management information in which a logical address indicating a slice of the master logical disk and the backup logical disk and a physical address of the disk device are associated, and the master logical disk and the backup logical disk are different A disk array control method used for a disk array device configured by a set of devices,
A failure detection step of detecting whether or not a failure has occurred in a disk device constituting the master logical disk;
When the master logical disk and the backup logical disk are in a synchronized state with respect to host access, if the failure detection step detects that a failure has occurred in a disk device constituting the master logical disk, the master logical disk Replacing a disk device in which a failure of the master logical disk and a disk device in which a failure of the backup logical disk has not occurred by replacing slices of a logical disk and the backup logical disk;
A state change step for changing the master logical disk and the backup logical disk to a disconnected state for host access when the replacement process is executed;
A disk array control method comprising:

Images