JPH10260788A - Disk array system - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To effectively use a hot spare disk. A disk having a larger storage capacity is used as a hot spare disk in a disk array. Controller 13-1,
13-2 is set as the storage area 112 of the upgrade controller F / W for upgrading the firmware required for the operation of the array controller 13-2, and the array controllers 13-1 and 13
-2, when the hot spare disk 110-6 is replaced,
The upgrade controller F / W stored in the F / W storage area 112 of the disk 110-6 is loaded into its own controller, and while one is being loaded, the other is loaded according to a request from the host device 12 while the disk is being loaded. Access control of the array 11 is performed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk array system having a plurality of disks, and more particularly to a disk array system suitable for effectively using a hot spare disk as a substitute disk and its mounting slot.

[0002]

2. Description of the Related Art A disk array system of this kind has an R
RAID (redundant arrays of in AID3)
Taking a disk array system having an expensive disk configuration as an example, as shown in FIG. 11, a plurality of data storage disks (data disks) in which data from the host device are dispersedly stored, for example, four data disks A storage disk 10-1 to 10-4, a disk (parity disk) 10-5 for storing parity data as error correction information of data stored in the disks 10-1 to 10-4,
The disk array 1 generally includes a disk (hot spare disk) 10-6 in place of the disks 10-1 to 10-5. Usually, the storage capacities of the disks (disk drives) 10-1 to 10-6 are set to be equal.

In the disk array system shown in FIG. 11, when a failure occurs in one of the disks 10-1 to 10-5, a process of restoring the data of the failed disk using the data of the remaining normal disks is performed. By storing the restored data in the hot spare disk 10-6, the hot spare disk 10-6 is used in place of the failed disk.

[0004]

As described above, in a conventional disk array system provided with a hot spare disk separately from a data disk and a parity disk,
Even if a failure occurs in the data disk or the parity disk, it is possible to cope without using the hot spare disk in place of the failed disk without stopping the system.

[0005] However, the hot spare disk is not used at all in a state where no failure occurs in other disks, and the mounting slot of the hot spare disk is not effectively used.

As described above, in the conventional disk array system, the hot spare disk is used only for the original purpose of the hot spare disk, and is not used at all in a normal state where no failure occurs in other disks. There is a problem that the mounting slot of the disk is wasted because it is not used effectively.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a disk array system that can effectively use a hot spare disk. Another object of the present invention is to provide a disk array system that can effectively use a mounting slot of a hot spare disk.

[0008]

A disk array system according to a first aspect of the present invention includes a plurality of first disks (disk drives), of which data is stored on disks other than a specific first disk. A first storage method (for example, RAID3) in which the error correction information of the data is arranged on the specific first disk in a distributed manner.
Storage method referred to as “RAID5”) or a second storage format (eg, a storage method referred to as RAID5) in which data and error correction information of the data are distributed and arranged on all of the plurality of first disks. The disk array to which one is applied and the disk array in accordance with a request from the host device are stored in a controller program (controller F /
Duplicated first for access control according to W)
And a second array controller, which is a part of the disk array and is used in place of the failed disk when a failure occurs in any one of the plurality of first disks. A data area for storing data of the failed disk, and an upgrade program (upgrade controller F / W) necessary for upgrading the controller program of the first and second array controllers are stored. Second having program storage area (F / W storage area)
The first and second array controllers have an F / F of a second disk (hot spare disk) in the disk array when the version of their controller program needs to be upgraded. An operation of loading the version upgrade controller F / W stored in the W storage area into the controller (own controller) is performed, and while one is being loaded, the other is loaded with the disk array in response to a request from the host device. It is characterized in that it is configured to perform access control.

In the disk array system having such a configuration, when a failure occurs in any one of the plurality of first disks, the failure is generated by one of the first and second array controllers. The stored contents of the failed disk are restored from the data and error correction information stored in the normal first disk other than the disk, and the restored contents are stored in the data of the hot spare disk (second disk). The fact that the hot spare disk can be used as a substitute disk for the failed disk by being stored in the area is no different from the conventional disk array system.

If only such a hot spare disk is used, the hot spare disk is not used at all in a normal state (a state in which no failure occurs in the first disk), so that the hot spare disk cannot be effectively used. It is useless.

However, in the disk array system having the above configuration, the storage capacity of the hot spare disk, in which the F / W storage area for storing the upgrade controller F / W is secured separately from the data area, that is, the first disk, has a larger storage capacity. A hot spare disk in which a large area having a large storage capacity is allocated as an F / W storage area is used, and by utilizing the fact that the array controller is duplicated, the controller F / W for the array controller is used. If the version of the hot spare disk needs to be upgraded, the array controller
An operation of loading the upgrade controller F / W stored in the W storage area into its own controller is performed, and during this loading, the other controller performs access control on the disk array in response to a request from the host device. Therefore, the version of the controller firmware of the array controller can be upgraded without using the hot spare disk (the F / W storage area therein) without stopping the system.

Here, as a condition for judging that the version of the controller F / W of the array controller needs to be upgraded, in addition to the fact that the hot spare disk has been replaced, the version upgrade of the first and second array controllers is performed. Execute (Upgrade controller F
/ W loading), in the F / W storage area of the hot spare disk, the loading of the upgrade controller F / W to the disk controller corresponding to the first and second array controllers, respectively. One of the first and second array controllers monitors the status of the hot spare disk. When the hot spare disk is replaced, one of the first and second array controllers has storage completion flag information indicating whether or not the completion has been completed. If the corresponding storage completion flag information indicates incomplete, loading of the version upgrade controller F / W is performed, and after the completion, the storage completion flag information is updated to the completion state, and the other array controller itself is The storage completion flag information corresponding to the If storage completion flag information indicates the completion, by performing the loading of version upgrade controller F / W, it may be configured to update the storage completion flag information corresponding to itself after completion completion state.

[0013] A disk array system according to a second aspect of the present invention includes various types of disks instead of the F / W storage area (program storage area) in the second disk in the disk array system according to the first aspect. In addition to providing a maintenance information storage area for storing maintenance information, when the first and second array controllers are not in an operation state in which they themselves control access to the disk array in response to a request from the host device, That is, when the system is in the standby state, it has a function of collecting various types of fault information in the system and statistical information about predetermined items and storing the statistical information in the maintenance information storage area of the second disk. Features.

In the disk array system having such a configuration, the hot spare disk in which the maintenance information storage area is secured separately from the data area, that is, the storage capacity of the first disk is larger than that of the first disk, and the area of the larger storage capacity is maintained. Using the hot spare disk allocated as the information storage area and utilizing the fact that the array controller is duplicated (by the first and second array controllers), the operating controller can issue a request from the host device. During the normal operation of controlling the access to the disk array according to the condition, the standby controller collects the failure information and the statistical information and stores them in the maintenance information storage area of the hot spare disk, so that the maintenance information is efficiently stored. Can be collected. In addition, hot spare disks that are not generally used in the normal state can be used without stopping the system.
In addition, since the hot spare disk can be removed from the system and replaced with another hot spare disk without affecting the host device connected to the system, maintenance information can be easily collected.

In the disk array system according to the third aspect of the present invention, it is necessary to upgrade the firmware for the controller in the F / W storage area of the second disk in the disk array system according to the first aspect. In place of the special version upgrade controller F / W, a version upgrade program (version upgrade disk F / W) necessary for version upgrade of the disk program (disk F / W) is stored, and the first and the above-mentioned first and second versions are stored. In the second array controller, the disk F / F for the first disk is
When it is necessary to upgrade the version of W, an operation of loading the version upgrade disk F / W stored in the F / W storage area of the second disk in the disk array into each of the first disks is performed. It is characterized by having a function of performing the order.

In the disk array system having such a configuration, the upgrade of the disk F / W of each of the first disks is stopped by effectively utilizing the hot spare disk (the F / W storage area therein). It is possible to do without doing.

Here, the first and second array controllers use the fact that the hot spare disk has been replaced as a criterion for determining that the version of the disk F / W of each of the first disks needs to be upgraded. Is when it is on standby (ie when it is not running)
The status of the hot spare disk is monitored, and if the hot spare disk is replaced, the first disk is detached from the system and the version upgrade disk F / W is detached from the system.
, The restoration of the data of the first disk, and the incorporation into the system may be sequentially performed. This data restoration of the first disk is necessary when a write request is issued from the host device while the disk is disconnected from the system.
It is restored in the same way as when a disk failure occurs.

[0018] A disk array system according to a fourth aspect of the present invention comprises a part of the disk array in the disk array in the disk array system according to the first aspect; When a failure occurs in any one of the first and second disks, one of the second disk used in place of the failed disk and the backup device used for backup of each of the first disks is selectively used. The second above
A common mounting slot is provided for the disk and the backup device, and the first and second array controllers monitor the mounting slot when they are in a standby state (that is, when they are not in an operating state), When detecting that the backup device is mounted, a function of copying the data of each of the first disks to the backup device is provided.

In the disk array system having such a configuration, a backup device (e.g., realized by a large-capacity storage medium device such as a magneto-optical disk device) having the same mounting structure as a hot spare disk (e.g., realized by a magnetic disk device). Is replaced with the hot spare disk and mounted in the corresponding mounting slot of the disk array, the fact is detected by the disk controller on the standby side, and the storage contents of the first disks in the disk array are stored in the backup device. Since the data is copied (saved), data backup from the disk array to the backup device can be performed without stopping the system.

[0020]

Embodiments of the present invention will be described below with reference to the drawings. [First Embodiment] FIG. 1 is a block diagram showing a configuration of a disk array system according to a first embodiment of the present invention.

The disk array system shown in FIG.
The AID3 comprises a disk array 11, a host device 12, and a dual array controller 13-1 (# 1), 13 for controlling access to the disk array 11 in response to a request from the host device 12. -2 (# 2).

The disk array 11 has a plurality of data storage disks in which data streams from the host device are dispersedly stored, for example, four data storage disks (hereinafter, referred to as data disks) 110-1 (#). 1) ~
110-4 (# 4); a disk (hereinafter, referred to as a parity disk) 110-5 for storing parity data as error correction information of data stored in the disks 110-1 to 110-4; Disks 110-1 to 110
-5 replacement disk (hereinafter, referred to as a hot spare disk) 110-6. In the present embodiment, each of the disks 110-1 to 110-6 is a hard disk (hard disk drive), that is, a magnetic disk (magnetic disk device).

The storage capacities of the disks 110-1 to 110-5 are equal, and the storage capacity of the disk (hot spare disk) 110-6 is larger than the storage capacities of the disks 110-1 to 110-5. Of the storage area of the disk 110-6, the area corresponding to the storage area of the disks 110-1 to 110-5 (the area equal to the storage capacity of the disks 110-1 to 110-5) is the disk 110-1 to 110-. 5 is assigned to the data storage area 111 for storing the data of the failed disk when a failure occurs in any one of
The remaining area is allocated to, for example, a firmware program storage area (hereinafter, referred to as a firmware storage area) 112. The F / W storage area 112 has a firmware upgrade F / W for the array controllers 13-1 and 13-2.
(Firmware) is stored in advance. Also, F /
At a predetermined position in the W storage area 112, a version upgrade F / W (control F / W) is
Storage completion flags 113-1 (F1), 113-2 indicating whether or not they have been stored (loaded) in 3-1 and 13-2.
(F2) is secured.

The array controllers 13-1 and 13-2 control access to the disk array 11 in accordance with a controller F / W stored in a program storage area (not shown) of the controllers 13-1 and 13-2. It has become. The array controllers 13-1 and 13-2 generate an exclusive OR circuit (EX-OR) 130 used for generating parity data and restoring data of a failed disk among the disks 110-1 to 110-5. -1, 130-2.

Next, a description will be given of a firmware upgrade process of the array controllers 13-1 and 13-2 in the configuration of FIG. First, the array controller 13-1, 1
If it is necessary to upgrade the firmware (controller F / W) used in 3-2, prepare a hot spare disk 110-6 storing the upgrade controller F / W in the F / W storage area 112. Then, the hot spare disk 110-6 is replaced with the current hot spare disk in the disk array 11. The system shown in FIG. 1 shows a state after the replacement of the hot spare disk. In this example, when it is necessary to upgrade the version of the controller F / W, the disk array 11
As shown in FIG. 2, a hot spare disk 110 having the same storage capacity as the disks 110-1 to 110-5 is mounted in the mounting slot of the hot spare disk. It is replaced with the prepared hot spare disk 110-6.

Now, the current hot spare disk (11
0) changes the version upgrade controller F / W to F /
Hot spare disk 11 stored in W storage area 112
When it is replaced with 0-6, for example, the array controller 13-1
(# 1) uses the upgrade controller F / W stored in the disk 110-6 (in the F / W storage area 112 therein) and uses its own controller F / W as indicated by reference numeral 31 in FIG. / W is upgraded.

While the firmware upgrade (31) is being executed by the array controller 13-1, the controller 13-1 cannot be used by the host device 12. However, in the present embodiment, since the array controller has a dual configuration of the array controller 13-1 and the array controller 13-2, the host device 1
No. 2 can access the disk array 11 as shown by reference numeral 32 in FIG. 3 using the array controller 13-2 (path on the side), so that there is no operational effect on the disk array system. When the firmware upgrade in the array controller 13-1 is completed,
The firmware upgrade in the array controller 13-2 is executed. In this case, the host device 12 accesses the disk array 11 using the array controller 13-1 (path on the side).

Hereinafter, the array controllers 13-1 and 13-2 will be described.
4 shows details of the F / W version upgrade process in FIG.
This will be described with reference to the flowchart of FIG. First, the array controller 13-1 (# 1) periodically patrols the status of the hot spare disk of the disk array 11 (step S1). Now, as shown in FIG. 2, the controller F / W for upgrading is replaced with a new hot spare disk 110-6 stored in the F / W storage area 112 from the current hot spare disk 110, and the system state as shown in FIG. As a result, the array controller 13-1 recognizes that the hot spare disk has been replaced (step S2).

When the array controller 13-1 recognizes that the hot spare disk has been replaced, the new hot spare disk, that is, the hot spare disk 110-6 has the F / W storage area 112, and
The storage completion flag 113-1 (F1) for the controller # 1 secured at a predetermined position in the / W storage area 112 is checked (step S3).

If the storage completion flag 113-1 for the controller # 1 is in the "not-yet" state (a state indicating that the version-up controller F / W has not been completely stored), the array controller 13-1 sets the hot spare disk. 110
-6, the version upgrade controller F / W stored in the F / W storage area 112 of its own controller F / W.
An operation of loading a W storage area (not shown) and upgrading the version of its own controller F / W is executed (steps S4 and S5).

When the loading of the upgrade controller F / W is completed, the array controller 13-1 sets the storage completion flag 113-1 for the controller # 1 in the F / W storage area 112 of the hot spare disk 110-6. Is set to the "finished" state (step S6), the version upgrade process of the controller F / W ends, and the process returns to the normal process (main routine).

As described above, the array controller 13
During the version upgrade process of the controller F / W at -1, the host device 12 accesses the disk array 11 using the other array controller 13-2 (path on the side).

On the other hand, the array controller 13-2 (# 1)
Also, similarly to the array controller 13-1, the status of the hot spare disk of the disk array 11 is periodically patroled (step S11). If the array controller 13-2 is replaced with a hot spare disk 110-6 as in this embodiment, the disk controller 13-2
The storage completion flag 113-1 (F1) for the controller # 1 secured at a predetermined position in the F / W storage area 112 of 0-6 is in the "completed" state, and the storage completion flag 113-113 for the controller # 2 is set. 2 (F2) is confirmed to be in the "not yet" state (steps S12 to S15), the version upgrade controller F / W stored in the F / W storage area 112 of the hot spare disk 110-6 is replaced by its own controller. A process of loading the firmware in the F / W storage area and upgrading the version of the controller F / W is executed (step S16).

When the loading of the upgrade controller F / W is completed, the array controller 13-1 sets the storage completion flag 113-2 for the controller # 2 in the F / W storage area 112 of the hot spare disk 110-6. (F2) is set to the "finished" state (step S1).
7), the version upgrade process of the controller F / W ends, and the process returns to the normal process (main routine).

During the version upgrade process of the controller F / W in the array controller 13-2, the host device 1
2 accesses the disk array 11 using the other array controller 13-1 (path on the side).

In the above, the free area of the hot spare disk (area other than the data storage area 111) is used as the storage area (F / W storage area) 112 for the control F / W for upgrading, and the array controller 13-
Although a case has been described where the control firmware is upgraded without stopping the system, the present invention is not limited to this. For example, it can be used as a storage area for maintenance information. [Second Embodiment] An embodiment of a disk array system using a hot spare disk having a maintenance information storage area in addition to a data storage area will be described with reference to the drawings.

FIG. 5 is a block diagram showing the configuration of the disk array system according to the second embodiment of the present invention. The disk array system shown in FIG. 5 includes a disk array 51 (corresponding to the disk array 11, host device 12, and array controllers 13-1 and 13-2 in FIG. 1), a host device 52, and an array controller 53-1 (# 1),
53-2 (# 2). Disk array 51
Has a disk configuration similar to that of the disk array 11 in FIG. 1, and is composed of four data disks, one parity disk (not shown), and one hot spare disk 510a. ing.

The hot spare disk 510a has a larger storage capacity than the data disk and the parity disk, and in this respect, the hot spare disk 1 in FIG.
Similar to 10-6. However, hot spare disk 5
In the storage area 10a, the area corresponding to the storage area of the data disk and the parity disk is allocated to the data storage area 511 (similar to the hot spare disk 110-6), but the remaining area is (F / W). The hot spare disk 110-6 is different from the hot spare disk 110-6 in that it is allocated to a storage area 512 for maintenance information (not a storage area).

In the system shown in FIG. 5, it is assumed that the array controller 53-1 of the array controllers 53-1 and 53-2 controls the access to the disk array 51 (on the operating side). In this case, the host device 52 uses the array controller 53-1 (path on the side) to access the disk array 51 as indicated by reference numeral 541 in FIG. During such normal operation,
When the other (standby side) array controller 53-2 detects any failure, the detected failure information is displayed in FIG.
As indicated by reference numeral 542 in FIG.
The hot spare disk is stored in the maintenance information storage area 512 of the disk 510a. The array controller 53-2 periodically collects various kinds of predetermined statistical information in the system,
In the maintenance information storage area 512.

When the maintenance worker needs to collect maintenance information (for example, at the time of periodic inspection or when trouble is required), the current hot spare disk 510a
Is replaced with another hot spare disk 510b as shown by reference numeral 543 in FIG. Thereby, the maintenance worker can collect the maintenance information from the maintenance information storage area 512 of the hot spare disk 510a.

In the first embodiment, the free area of the hot spare disk is used as a storage area (F / W storage area) for the control F / W for upgrading, and the control F / W of the array controller is used. The disk array system that can be upgraded has been described.
F / W for operation control on data disk (disk drive) and parity disk (disk drive)
(Disk F / W) can also be used for upgrading. [Third Embodiment] An embodiment of a disk array system using a hot spare disk having a disk F / W storage area in addition to a data storage area will be described with reference to the drawings.

FIG. 6 is a block diagram showing the configuration of the disk array system according to the third embodiment of the present invention. The disk array system in FIG. 6 includes a disk array 61 (corresponding to the disk array 11, host device 12, and array controllers 13-1 and 13-2 in FIG. 1), a host device 62, and an array controller 63-1 (# 1), 6
3-2 (# 2). The disk array 51
The disk configuration is the same as that of the disk array 11 in FIG. 1, and four data disks (disk drives)
610-1 (# 1) to 610-4 (# 4), one parity disk (disk drive) 610-5 (# 5),
One hot spare disk (disk drive) 61
0-6.

The hot spare disk 610-6 includes the data disks 610-1 to 610-4 and the parity disk 61.
It has a storage capacity larger than 0-5, and among the storage areas, an area corresponding to the storage areas of the data disks 610-1 to 610-4 and the parity disk 610-5 is left in the data storage area 611. Area is the F / W storage area 61
2 is similar to the hot spare disk 110-6 in FIG. However, the F / W storage area 612 of the disk 610-6 has data disks 610-1 to 610-4 and a parity disk 610-5.
F / W (disk F / W) for version upgrade is stored, which is different from the disk 110-6 in this point.
At a predetermined position of the F / W storage area 612 of the disk 610-6, a version upgrade F / W (disk F / W)
Is a disk (disk drive) 610-1 to 610-5
(Storage completion flag) 613-1 indicating whether or not the data has already been stored (loaded) in (disk F / W storage area).
(F1) to 613-5 (F5) are secured.

The array controllers 63-1 and 63-2 control access to the disk array 11 in accordance with a controller F / W stored in a program storage area (not shown) of the controllers 63-1 and 63-2. It has become. The array controllers 63-1 and 63-2 generate an exclusive OR circuit (EX-OR) 630 used for generating parity data and restoring data of a failed disk among the disks 610-1 to 610-5. -1, 630-2.

Next, the disk 610 in the configuration of FIG.
The F / W version upgrade process of -1 to 610-5 will be described with reference to the flowchart of FIG. First,
When it is necessary to upgrade the F / W (disk F / W) used in the disks 610-1 to 610-5, the hot spare disk storing the upgraded disk F / W in the F / W storage area 612 610-6 is prepared, and the hot spare disk 610-6 is replaced with the current hot spare disk in the disk array 61. The system shown in FIG. 6 shows a state after the replacement of the hot spare disk. Here, the hot spare disk before replacement with the hot spare disk 610-6 is a hot spare disk having only a data storage area or a hot spare disk similar to the hot spare disk 610-6 replaced last time.

Now, the current hot spare disk stores the upgrade disk F / W in the F / W storage area 61.
It is assumed that the hot spare disk 610-6 stored in No. 2 has been replaced. At this time, the array controllers 63-1 and
For example, the array controller 63-1 (#
It is assumed that 1) (on the operating side) controls access to the disk array 61.

Array controller (standby side) 63-2
(# 2) is the array controller 63-1 (on the operation side)
If (# 1) is in charge of access control, the status of the hot spare disk of the disk array 61 is periodically patroled (step S21). Now, upgrade the upgrade disk F / W from the current hot spare disk to F
Assuming that the system has been replaced with a new hot spare disk 610-6 stored in the / W storage area 612 and the system has entered the system state as shown in FIG.
It recognizes that the hot spare disk has been replaced (step S22).

When the array controller 63-2 recognizes that the hot spare disk has been replaced, the array controller 63-2 determines that the new hot spare disk, that is, the hot spare disk 610-6 has the F / W storage area 612, and
Disk 610-1 (# 1) using the upgrade disk F / W stored in the / W storage area 112.
The processing for sequentially upgrading the disk F / Ws of .about.610-5 (# 5) is executed as follows.

First, the array controller 63-2 controls the disk 61 to be upgraded by the disk F / W.
A variable (pointer) n specifying 0-n (disk #n) is set to an initial value "1" (step S23).

Next, the array controller 63-2 operates the disk 610-n (disk #) secured at a predetermined position in the F / W storage area 612 of the hot spare disk 610-6.
The storage completion flag 613-n for n) is checked (step S24). Here, since n = 1, the storage completion flag 613-1 (F1) for the data disk 610-1 (disk # 1) is checked (step S2).
4).

If the storage completion flag 613-1 (F1) for the data disk 610-1 (disk # 1) is in the "not yet" state (a state indicating that the storage of the version upgrade disk F / W is not completed). If (step S25), the array controller 63-2 disconnects the data disk 610-1 (disk # 1) from the system (step S26). Here, disconnecting a disk means that actual access is not performed in response to an access request from the host device 62.

In this state, the array controller 63-2
F / W storage area 612 of hot spare disk 610-6
Upgrade disk F / W stored in the data disk 61 (separated from the system)
The data disk 610-1 (# 1) is loaded on the disk F / W storage area (not shown) of the data disk 610-1 (# 1).
The operation of upgrading the version of the disk F / W is executed (step S27).

Now, the disk F / W of the data disk 610-1 (# 1) (by the array controller 63-2)
During the period in which is upgraded, the data disk 610-1 (# 1) cannot be used. Therefore,
During this period, the host device 62 sends the array controller 63
When the access to the disk array 61 (disk read access or disk write access) is requested to the RAID controller -1, the array controller 63-1 performs the operation in the RAID 3 configuration described below by the disk read access or disk write access. This is performed according to which of the write access.

First, the disk read operation during the F / W version upgrade of the disk 610-1 (# 1) will be described with reference to FIG. In the disk array system having the RAID3 configuration, the data stream from the host device 62 is divided into blocks (chunks) of a fixed size S, and the data disks 610-1 (# 1) to 610-4 (# 4).
Are distributed. Here, disks 610-1 to 610-4
If the four consecutive blocks distributed and arranged in the same manner are di1 to di4, the parity data pi which is the exclusive OR result of the blocks di1 to di4 at the same bit position is allocated to the parity disk 610-5 (# 5). . These di1 to di4,
Generally, pi is arranged in the same disk area of each of the disks 610-1 to 610-4 and 610-5.

During the firmware upgrade of the disk 610-1 (# 1), the disk 610-1 (# 1) cannot be accessed. Therefore, F of disk 610-1 (# 1)
During the / W version upgrade, the relevant disk 610-1 (#
For example, if a read (disk read) of the data di1 stored in 1) is requested from the host device 62, the array controller 63-1 sends the data disks 610-2 (# 2) to 610-4. The corresponding data di2 to di4 in (# 4) and the parity disk 610-5 (#
5) The corresponding data (parity data) pi in (5) is read out in parallel, and an exclusive OR circuit (EX-OR) 630 is read out.
At -1, the exclusive OR of these data di2 to di4, pi is obtained for each same bit position. This exclusive OR circuit (EX
The result of the exclusive OR of the data di2 to di4, pi according to (-OR) 630-1 matches the data di1 requested by the host device 62. Therefore, the array controller 63
-1 indicates the exclusive OR result of the data di2 to di4 and pi output from the exclusive OR circuit (EX-OR) 630-1 as read data 80 requested by the host device 62, send.

Next, the F / W of the disk 610-1 (# 1)
Fig. 9 shows the disk write operation during version upgrade.
This will be described with reference to FIG. F / of disk 610-1 (# 1)
During the W version upgrade, the relevant disk 610-1 (#
1) cannot be accessed. Therefore, the disk 610-1
During the firmware upgrade of (# 1), if at least a disk write operation involving data writing to the disk 610-1 (# 1) is requested from the host device 62, the array controller 63-1 will execute Is performed.

First, it is assumed that write data 90 (a stream thereof) accompanying a disk write request from the host device 62 has four blocks distributed and arranged on the data disks 610-1 to 610-4, and the four blocks are referred to as di1 to di1. di4. In this case, the array controller 63-1 divides (stripes) the write data 90 into blocks di1 to di4. Then, the array controller 63-1 obtains the parity data pi by taking the exclusive OR of each of the same bit positions of the blocks di1 to di4 in the exclusive OR circuit 630-1, and obtains the parity data pi. Block d in blocks di1 to di4 is stored on disk 610-5.
i2 to di4 are written in parallel to the data disks 610-2 to 610-4. If the write data 90 consists of only the block di1, for example, the blocks di2 to di4 corresponding to the block di1 are stored in the data disks 610-2 to 610.
0-4, the parity data pi of the blocks di1 to di4 is generated, and the parity data pi may be written to the parity disk 610-5.

When the firmware upgrade (step S27) of the disk 610-1 (# 1) is completed, the array controller 63-1 updates the disk 610-1 (# 1).
1) Restoring the data, that is, because the disk 610-1 (# 1) is in the process of upgrading the firmware, the disk 610-1
Data that could not be written to (# 1) is restored (step S28). This disk 610-1
The data restoration of (# 1) is performed on the data disk 610-2.
(# 2) to 610-4 (# 4) and parity disk 610
-5 (# 5) data can be obtained by taking the exclusive OR of the same bit positions for each block. For example, the data block to be restored is di1, the block d
Data disk 610-2 (# 2) to 610 corresponding to i1
Assuming that the data block of -4 (# 4) is di2 to di4 and the data block (parity data) of the parity disk 610-5 (# 5) is pi, di1 is di2 to di4, pi.
The exclusive OR circuit (EX-OR) 630-1 obtains the exclusive OR of each of the same bit positions to be restored.

The array controller 63-1 has a disk 6
When the data restoration of 10-1 (# 1) (that is, the data restoration of the disk indicated by the current pointer n) is completed, the disk 610- secured at a predetermined position in the F / W storage area 612 of the hot spare disk 610-6. 1 (disk #
The storage completion flag 613-18 (F1) for 1) is set to the "completed" state (a state indicating the completion of storage of the upgrade disk F / W) (step S29).

Next, the array controller 63-1 incorporates the disk 610-1 (# 1) into the system again (step S30), and the array controller 63-1 upgrades the version of the disk F / W as in this example. Is not completed (step S31), the pointer n is incremented by 1 to designate a new disk 610-n (here, the data disk 610-2) to be upgraded in the disk F / W. And this new disk 61
0-n, that is, data disk 610-2 (# 2)
As with disk 610-1, disk F
The processing after step S24 for upgrading / W is executed.

When the version upgrade of the disk F / W for the data disk 610-2 (# 2) is completed, the array controller 63-1 completes the remaining data disks 610-2.
-3 (# 3), 610-4 (# 4) and parity disk 6
Also for 10-4 (# 5), in the same manner as for the disk 610-1, the processing after step S24 for upgrading the disk F / W is executed to configure the disk array 61 (hot spare disk). 610
Discs 610-1 (# 1) to 610-5 (#
5) F / W version upgrade is completed. [Fourth Embodiment] Next, a fourth embodiment of the present invention will be described with reference to the drawings.

FIG. 10 is a block diagram showing the configuration of the disk array system according to the fourth embodiment of the present invention.
FIG. 10 shows that the disk array system is a disk array 10 (corresponding to the disk array 11, host device 12, and array controllers 13-1 and 13-2 in FIG. 1).
1. Host device 102 and array controller 103-1
(# 1) and 103-2 (# 2). The disk array 101 has the same disk configuration as the disk array 11 in FIG. 1, and has four data disks (disk drives) 1010-1 (# 1) to 1010-4 (#
4) and one parity disk (disk drive)
This shows a state composed of 1010-5 (# 5) and one hot spare disk (disk drive) 1010-6.

In the system shown in FIG. 10, when the hot spare disk 1010-6 is removed from its mounting slot, the disk array 10
A backup device (backup medium) 1010 used for backing up data of the data disk 1 (the data disks 1010-1 to 1010-4) can be mounted. However, the form factor (mounting structure) of the backup device 1010 is the hot spare disk 1
It must be the same as that of 010-6. As the backup device 1010, a large-capacity recording media device such as a magneto-optical disk device having a capacity capable of backing up data of the disk array 101 (the data disks 1010-1 to 1010-4 therein) is used.

Now, the system maintenance staff is in charge of the disk array 101 (the data disks 1010-1 to 1010 in the disk array 101).
To back up the data of -4), the hot spare disk 1010-6 mounted in the mounting slot of the hot spare disk of the disk array 101 is required.
Is replaced with a backup device 1010 (having the same form factor as the hot spare disk 1010-6) as shown by the arrow 104. That is, the backup device 1010 is mounted in the mounting slot instead of the hot spare disk 1010-6.

Now, the array controllers 103-1 and 103
-2 of array controllers 103-2 (# 2)
Manages access control of the disk array 101.

The array controller 103-1 (# 1)
When the array controller 103-2 (# 2) is in charge of access control, the status of the hot spare disk of the disk array 101 is regularly patroled. When the array controller 103-1 recognizes that a device other than the hard disk device (magnetic disk device) is mounted in the mounting slot of the hot spare disk by this patrol, the backup device 1010 such as a magneto-optical disk device or the like.
Is determined to have been implemented.

Then, the array controller 103-1 operates the data disks 1010-1 to 1010-1 to 1 in the disk array 101.
A backup (copy) process of reading out the storage data of 010-4 as indicated by reference numeral 105 and storing it in the backup device 1010 is performed.

When the above-described backup processing is completed, the array controller 103-1 notifies the maintenance staff of the completion by a display output or the like. As a result, the maintenance person can replace the backup device 1010 with the hot spare disk 101.
Replace with 0-6 to restore the original system configuration.

The case where the present invention is applied to a disk array system having a RAID3 configuration has been described above.
The present invention is also applicable to disk array systems other than the RAID3 configuration, such as a disk array system having a RAID5 configuration. In a disk array system having a RAID 5 configuration, a dedicated parity disk is not provided, and parity data is distributed and arranged on each data disk similarly to data (data stream).

[0070]

As described above in detail, according to the present invention, the version upgrade of the controller program of the array controller is performed by using the dual array controller and the replacement second disk (hot spare disk). This can be performed without stopping the system by effectively using the version upgrade controller program stored in the program storage area secured in the parentheses.

Further, according to the present invention, the system maintenance information can be stored without stopping the system by effectively using the redundant array controller and the maintenance information storage area secured in the hot spare disk. The second disk for which maintenance information has been collected can be exchanged without stopping the system.

Further, according to the present invention, the version upgrade of the program of each disk (first disk) in the disk array is performed in the dual storage of the array controller and in the program storage area secured in the hot spare disk. The stored version upgrade disk program can be used effectively without stopping the system.

According to the present invention, a backup device realized by a large-capacity recording media device such as a magneto-optical disk device having the same mounting structure as a hot spare disk is replaced with the hot spare disk and a corresponding mounting slot of the disk array is replaced. By using the fact that the array controller is duplicated, the storage contents of each disk (first disk) in the disk array are automatically saved in the backup device without stopping the system. can do.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a disk array system according to a first embodiment of the present invention.

FIG. 2 is a diagram showing a state in which an old hot spare disk is replaced with a new hot spare disk storing a version upgrade controller F / W in the embodiment.

FIG. 3 shows an array controller 13 according to the embodiment.
FIG. 7 is a diagram showing a state in which the version upgrade of the controller F / W by -1 and the disk access by the array controller 13-2 are performed in parallel.

FIG. 4 is an exemplary flowchart for explaining version upgrade processing of the controller F / W in the embodiment.

FIG. 5 is a block diagram showing a configuration of a disk array system according to a second embodiment of the present invention.

FIG. 6 is a block diagram showing a configuration of a disk array system according to a third embodiment of the present invention.

FIG. 7 is a flowchart for explaining version upgrade processing of a disk F / W in the configuration of FIG. 6;

FIG. 8 shows a disc 610-1 (# 1) in the configuration of FIG.
FIG. 9 is a diagram for explaining a disk read operation during F / W version upgrade.

FIG. 9 shows a disk 610-1 (# 1) in the configuration of FIG.
FIG. 7 is a diagram for explaining a disk write operation during F / W version upgrade.

FIG. 10 is a block diagram showing a configuration of a disk array system according to a fourth embodiment of the present invention.

FIG. 11 is a diagram showing a typical disk configuration example of a conventional disk array.

[Explanation of symbols]

11, 51, 61, 101: disk array, 12, 52, 62, 102: host device, 13-1, 13-2, 53-1, 53-2, 63-1, 63-2, 1
03-1, 103-2 ... array controller, 110-1 to 110-4, 610-1 to 610-4, 1010-1
... 1010-4 ... data disk (first disk), 110-5, 610-5, 1010-5 ... parity disk (first disk, specific first disk), 110, 110-6, 510a, 510b, 610-6,1
010-6: Hot spare disk (second disk), 111, 511, 611: Data storage area, 112, 612: F / W storage area, 512: Maintenance information storage area, 1010: Backup device.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor, Kyoichi Sasamoto 1 Toshiba-cho, Fuchu-shi, Tokyo Inside the Fuchu factory, Toshiba Corporation

Claims (6)

[Claims] 1. A data storage system comprising a plurality of first disks, wherein data is distributed and arranged on disks other than the specific first disk among the plurality of first disks, and data is distributed to the specific first disk. Either the first storage method in which the error correction information of the data is arranged, or the second storage format in which the error correction information of the data is distributed and arranged on all of the plurality of first disks, A duplicated first and second array controller for controlling access to the disk array according to a controller program in response to a request from a host device, and a part of the disk array. None, if any one of the plurality of first disks fails,
A data area for storing the data of the failed disk and a version upgrade program required for upgrading the controller program of the first and second array controllers are used in place of the failed disk. A second disk having a program storage area to be stored, wherein the first and second array controllers are provided with a second disk in the disk array when the controller program needs to be upgraded. An operation of loading the version upgrade program stored in the program storage area of the disk into the controller is performed, and while one is being loaded, the other controls access to the disk array in response to a request from the host device. Data Disk array system. 2. In the program storage area of the second disk, whether or not loading of a version-up program to the disk controller has been completed corresponding to the first and second array controllers, respectively. One of the first and second array controllers monitors the state of the second disk, and when the second disk has been replaced, If the storage completion flag information corresponding to itself indicates incomplete, loading of the version upgrade program is performed, and after completion, the storage completion flag information is updated to a completed state, and the first and second programs are updated. The other of the array controllers monitors the state of the second disk, and the storage completion flag information corresponding to the other disk controller indicates incomplete. If the storage completion flag information corresponding to the other array controller indicates completion, the version upgrade program is loaded, and after completion, the storage completion flag information corresponding to itself is updated to a completed state. 2. The method according to claim 1, wherein
The disk array system according to the above. 3. A method including a plurality of first disks, wherein data is distributed and arranged on disks other than the specific first disk among the plurality of first disks, and data is distributed to the specific first disk. Either a first storage method in which the error correction information of the data is arranged, or a second storage format in which the error correction information of the data is distributed and arranged on all of the plurality of first disks, A disk array to be applied; a duplicated first and second array controller for controlling access to the disk array in response to a request from a host device; If any one of the first disks fails,
A second disk having a data area for storing data of the failed disk and a maintenance information storage area for storing various types of maintenance information, the second disk being used in place of the failed disk. During a normal operation in which one of the first and second disk controllers controls access to the disk array in response to a request from the host device, various types of failure information in the system and statistics on predetermined items are provided. A disk array system, wherein information is collected and stored in the maintenance information storage area of the second disk. 4. A disk program including a plurality of first disks operating according to a disk program, wherein data is distributed and arranged on disks other than the specific first disk among the plurality of first disks, and A first storage method in which the error correction information of the data is arranged on the first disk, or a second storage method in which the data and the error correction information of the data are distributed and arranged on all of the plurality of first disks.
A disk array to which one of the storage formats is applied, a duplicated first and second array controller for controlling access to the disk array in response to a request from a host device, and the disk array And if any one of the plurality of first disks fails,
A data area for storing data of the failed disk, which is used in place of the failed disk, and a program for storing an upgrade program required for upgrading the disk program of the first disk A second disk having a storage area, wherein one of the first and second array controllers is controlling access to the disk array in response to a request from the host device. When it is necessary to upgrade the disk program for the first disk, the second disk in the disk array is updated.
Loading the version-up program stored in the program storage area of the first disk into each of the first disks in order. 5. One of the first and second array controllers monitors the status of the second disk, and when the second disk is replaced, targets each of the first disks. Wherein the disconnection from the system, the loading of the version upgrade program in a state disconnected from the system, the restoration of the data of the first disk, and the incorporation into the system are sequentially executed. Item 6. The disk array system according to Item 4. 6. A method including a plurality of first disks, wherein data is distributed and arranged on disks other than the specific first disk among the plurality of first disks, and data is distributed to the specific first disk. Either the first storage method in which the error correction information of the data is arranged, or the second storage format in which the error correction information of the data is distributed and arranged on all of the plurality of first disks, A disk array to be applied; a duplicated first and second array controller for controlling access to the disk array in response to a request from a host device; If any one of the first disks fails,
In the second disk and the backup device, one of a second disk used in place of the failed disk and a backup device used for backup of each of the first disks is selectively mounted. A common mounting slot, and one of the first and second array controllers monitors the mounting slot in a state where the other is controlling access to the disk array in response to a request from the host device. And detecting that the backup device is mounted, and copying the data of each of the first disks to the backup device.