JPH09305329A - Storage device array system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To clearly discriminate a storage device where any one of fault generating state, data reconstructing state and detachable state occurs, when at least one of such a state is detected, by reporting the detectable states of respective storage devices through a reporting means provided corresponding to the respective storage devices to the outside. SOLUTION: In the middle of basic operation of a hard disk array 10, three states of fault generating state, data reconstructing state and detachable state are detected by a central control unit(CCU) 71 of a main computer 70. Then, the CCU 71 sends out the signal of LED control through an input/output interface 90 to a controller 80. The controller 80 receives this signal and turns on any correspondent LED among 1st LED 30a-30d, 2nd LED 40a-40d and 3rd LED 50a-50d based on this signal. Therefore, a user can easily discriminate which storage device's state is the state detected and reported.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION A storage device array composed of a plurality of storage devices and capable of restoring the contents of the storage device in which the failure has occurred from the remaining storage devices when the storage device fails, and the storage device thereof. The present invention relates to a storage device array system including a main computer that controls an array.

[0002]

2. Description of the Related Art
As a storage device that enables high-speed access and guarantees high reliability, RAID (Redundant
Storage arrays represented by ant Arrays of Inexpensive Disks have been proposed. When storing data, the storage array also stores data protection information generated from the data. When a failure occurs in one of the plurality of storage devices, the storage device can be replaced with another storage device while the storage device array is operating, and the failed storage device can be replaced. The data stored in the storage device other than the device and the data protection information can be used to reconstruct the data stored in the failed storage device. Further, for maintenance, the storage device can be temporarily removed while the storage device array is operating.

As described above, in each of the storage devices constituting the storage device array, for example, a state in which the above-mentioned failure has occurred, a state in which data is being reconstructed, a state in which data can be replaced, or the like can occur. However, the storage device itself that constitutes the conventional storage device array is provided only with means for externally notifying the ON / OFF state of the power supply and the state of whether or not it is being accessed. The above-mentioned three states occurring in the storage devices constituting the storage device array were displayed on the display of the main computer controlling the storage device array, but could not be known from the storage device array itself. It was not possible to know about other operational statuses (data rebuilding status, exchangeable status, etc.).

When a failure occurs in the storage device,
The user had to replace the failed storage device based on the information displayed on the display. As a result, it is possible to accidentally replace a good storage device instead of a failed storage device, and if you make a wrong replacement, all the data stored in the storage device array is lost. , There was a risk of causing malfunction in the controlling main computer. Also,
The appearance of the storage device cannot be used to judge whether the data is being reconstructed or simply being accessed, and the user did not know whether the data was being reconstructed normally. Furthermore, when a storage device is removed for maintenance, it cannot be determined whether it is in a replaceable state, and if the storage device is accessed during the removal, the data in the storage device may be destroyed. was there. Also,
If you are unlucky, the storage device itself may break down.

The present invention has been made to solve the above-mentioned problems, and when at least one of the three states of the failure occurrence state, the data reconstruction state, and the removable state is detected, they are detected. It is an object of the present invention to provide a storage device array system capable of clearly discriminating in which storage device the above state occurs.

[0006]

Means for Solving the Problems and Effects of the Invention
The storage device array system according to claim 1, comprising a plurality of storage devices for storing data, and at least one storage device for storing data protection information generated from the data,
When a failure occurs in one of the storage devices, it is possible to reconstruct the data stored in the failed storage device from the remaining storage devices. In a storage device array system comprising a storage device array from which one storage device can be removed and a main computer for controlling the storage device array, the main computer has a failure occurrence state and a data reconstruction state of each storage device. At least one of the three removable states can be detected, and the state of each detectable storage device is notified to the outside by a notification means provided corresponding to each storage device. Is characterized by.

According to the storage device array system of the present invention, the main computer that controls the entire system detects at least one of the three states of the storage device: failure occurrence state, data reconstruction state, and removable state. can do. When the main computer detects the detectable states of the respective storage devices, the main computer informs the detected states of the respective storage devices to the outside by using the notifying means provided corresponding to the respective storage devices. In this way, since the detected state of each storage device is notified to the outside by the notification means provided corresponding to each storage device, the user can determine which storage device the detected and notified state is. It is possible to easily determine whether or not the state is.

The phrase "provided for each storage device" may mean that each storage device itself has an informing means, as described in claim 2, or the storage device itself. Instead, it is conceivable that the housing of the storage device array accommodating each storage device is provided at a position corresponding to each storage device, or the storage device itself has a means for notifying a certain state. It is conceivable that the means for notifying another state is provided at a position where the housing of the storage device array can be associated with each storage device.

By providing each storage device itself with the notification means, the user can more clearly discriminate the notified storage device in the above state. Further, in the case of detecting and notifying the occurrence state of the fault among the above three states, the storage device in which the fault has occurred can be easily determined. As a result, if you remove and replace a failed storage device, you are less likely to accidentally remove a good storage device, resulting in loss of data integrity or storage of the entire storage array system resulting from incorrect removal. It is possible to prevent the main computer that controls the device array from malfunctioning.

By detecting and notifying the data reconstruction state among the above three states, it is possible to distinguish between the data reconstruction and the simple access. Further, if the removable state of the above three states is detected and notified, it is possible to easily determine whether or not the removable storage device is in a removable state when one storage device is removed for maintenance or the like. Therefore, it is less likely that the storage device being accessed is removed and the data stored in the storage device is destroyed or the storage device itself is broken. Further, when one storage device is removed, there is no notice of the removable state, so that the mistake of removing another storage device and losing the data validity of the entire storage device array system is extremely small.

As the "informing means", an LED which has been conventionally used for informing ON / OFF of a power source and whether or not an access is being made may be prepared for each of the above three states.
LEDs of different colors may be prepared so that the above three states can be distinguished. Alternatively, it is conceivable that a liquid crystal display is used to display characters and figures according to the above three states.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram when the storage device array system of the present invention is applied to a hard disk array system 1.

The hard disk array system 1 includes a hard disk array 10 as a "storage device array",
The hard disk array 10 is composed of a main computer 70 as a "main computer". The main computer 70 includes a central controller 71 and a display 73. The central control device 71 is a well-known device including a CPU as a control unit, a ROM as a storage unit, a RAM as a temporary storage unit, and the like.

Hard disk array 10 of the present embodiment
Are four hard disk units 60a to 60d and "hard disk units 60a to 6" as "storage devices".
The controller 80 is provided with a controller 80 that controls reading and writing of data from and to 0d, and an input / output interface 90 that is connected to the controller 80 and inputs and outputs data to and from the main computer 70. In order to distinguish each hard disk unit 60a-60d, in the following description, A hard disk unit 60a, B hard disk unit 60b, C hard disk unit 60c,
It is described as a D hard disk unit 60d.

The A hard disk unit 60a is a hard disk 20 including a drive mechanism and a disk.
a, the first LED 30a as the "notifying means", the second L
The ED 40a and the third LED 50a are provided. 1st LE
The D30a is means for notifying the "fault occurrence state", the second LED 40a is means for notifying the "data reconstruction state", and the third LED 50a is means for notifying the "removable state".

The B to C hard disk unit 60
Since b to 60d have the same configuration as the A hard disk unit 60a, the description thereof will be omitted. Then, in the following description, the B to C hard disk unit 60
When describing the components b to 60d, the symbols of the respective components used in the A hard disk unit 60a are distinguished by adding lowercase letters b to d.

As described above, in the hard disk units 60a-60d of A to D of this embodiment, the above-mentioned three states, "failure occurrence state", and "data recovery" are added to the hard disk units 60a to 60d of A to D themselves. 1st, 2nd, 3rd L for informing "construction state", "removable state"
EDs are provided corresponding to the respective states.

For example, A hard disk unit 60a
When a failure occurs in the corresponding first LED 30a
Lights up. When the B hard disk unit 60b is in the data reconstruction state, the corresponding second LED 4
0b lights up. When the C and D hard disk units 60c and 60d are in the removable state, the corresponding third LEDs 50c and 50d are turned on.

FIG. 2 is an external view showing the external appearance of the hard disk array 10. 1st LED30a-30d, 2nd
LED 40a-40d, 3rd LED50a-50d is A
To D hard disk units 60a-60
It is attached to d itself so that it can be recognized from the outside.

The hard disk array 10 is RAI.
It constitutes D4. Here, the D hard disk unit 60d is assigned to the parity disk, and A to C
The hard disk units 60a-60c are assigned to the data disks. Here, the basic operation of the hard disk array 10 constituting the RAID 4 will be briefly described. In the hard disk array 10, the stored data is divided into a plurality of bytes. The divided data is assigned as a data disk A ~
It is stored in the C disk units 60a-60c. Then, parity data as "data protection information" is calculated based on the data stored in the A to C hard disk units 60a to 60c, and is stored in the D hard disk unit 60d assigned as the parity disk. As a result, the hard disk unit 60
When a failure occurs in one of a to 60d, it is possible to replace the failed hard disk unit and reconstruct the data stored in that hard disk unit from the data stored in the remaining hard disk units. it can.

For example, B hard disk unit 60b
If a failure occurs, the B hard disk unit 60b is removed and replaced with a new hard disk unit. The data stored in the B hard disk unit 60b is transferred to the new hard disk unit A, C, D hard disk units 60a, 60.
c, 60d can be reconstructed.

For maintenance purposes, one of the A to D hard disk units 60a to 60d can be removed while the power is on. In the present hard disk array system 1, the central controller 71 of the main computer 70 controls the "failure occurrence status", "data reconstruction status", and "removable status" during the basic operation of the hard disk array 10. Detect one condition. Then, the central control unit 71 sends an LED control signal to the controller 80 via the input / output interface 90. The controller 80 receives this signal, and based on this signal, the first LEDs 30a to 30d and the second LE
D40a to 40d and corresponding ones of the third LEDs 50a to 50d are turned on.

It is a feature of the present invention that the above-mentioned three states, "fault occurrence state", "data reconstruction state", and "removable state" are informed, which points will be described below. First, "failure occurrence state", which is one of the above three states
The notification will be described. In order to facilitate understanding of the description here, conventional notification of a failure occurrence state will be described first.

Conventionally, when a failure occurs in a certain hard disk unit, the central control unit 71 detects the failure occurrence and displays on the display 73 which hard disk unit among the A to D hard disk units 60a to 60d the failure has occurred. Was there. For example, when a failure occurs in the B hard disk unit 60b, the display 73
"B" was displayed above. Then, at this time, the user installs the B hard disk unit 60b corresponding to the display "B" on the display 73 in the A to D hard disk units 60a to 60d installed in the housing of the hard disk array 10 as described above. It was identified and exchanged.

However, there is a possibility that the user may mistake the display "B" on the display 73 and the B hard disk unit 60b for himself. Then, when a normal hard disk unit is replaced by mistake, the validity of the data of the entire hard disk array 10 is lost, or the main computer 70 malfunctions due to the loss of the validity of the data. There was a point.

On the other hand, in the hard disk array system 1 of the present embodiment, the central control unit 71 of the main computer 70 detects the failure occurrence state of each of the hard disk units 60a to 60d of A to D, and based on the detection result. The first LED 30 by transmitting a signal
The corresponding one of a to 30d is turned on.

For example, when the central controller 71 detects a failure occurrence state of the B hard disk unit 60b and sends an LED control signal to the controller 80, the controller 80 causes the first LED 30b provided corresponding to the B hard disk unit 60b to be turned on. Turn on the light. 1st LED3
Since 0b is provided in the B hard disk unit 60b itself, the user can easily recognize that a failure has occurred in the B hard disk unit 60b without performing the conventional correspondence by himself. As a result, it is extremely rare to mistakenly replace a normal hard disk unit, which prevents the data integrity of the entire hard disk array 10 from being lost and prevents the main computer 70 from malfunctioning due to the loss of the data integrity. be able to.

Next, the "data reconstruction state" of the above three states will be described. In the conventional system, only the access is informed, and the data reconstruction state is not detected and informed. When reconstructing data, all hard disk units 60a to 60d
60d is accessed. At this time, the user could not determine whether the user is simply accessing the data or rebuilding the data.

On the other hand, in the hard disk array system 1 of the present embodiment, the central control unit 71 of the main computer 70 detects the data reconstruction state, and the second hard disk unit is installed in the A to D hard disk units 60a to 60d.
The LEDs 40a to 40d inform the outside of which hard disk unit is reconstructing data.

For example, B hard disk unit 60b
On the other hand, when the data is being reconstructed, the central control unit 71 of the main computer 70 detects the data reconstructing state, and sends the LED control signal to the controller 80. As a result, the controller 80 turns on the second LED 40b provided on the B hard disk unit itself. Thereby, the user can easily recognize that the data is being reconstructed in the B hard disk unit 60b.

Next, the "removable state" of the above three states will be described. Similar to the data reconstruction state, this detachable state has not been previously notified. Therefore, if the hard disk unit is accessed during removal, the stored data may be destroyed, or the hard disk unit itself may fail.

On the other hand, in the hard disk array system 1 of this embodiment, when the central control unit 71 of the main computer 70 detects the removable state,
An LED control signal is sent to the controller 80 based on the detection result. As a result, based on the control signal, the controller 80 causes the third LEDs 50a to 5 provided on the hard disk units 60a to 60d of A to D themselves.
The corresponding one of 0d is turned on. This allows the user to easily determine which of the A to D hard disk units 60a to 60d is removable. As a result, it is extremely unlikely that the hard disk unit will be removed during access and the stored data will be destroyed, or that the hard disk unit itself will fail.

The "removable state" in the present embodiment means that there is no access and that none of the hard disk units 60a-60d of A to D has been removed. This is because, since parity data is used as the data protection information, if two hard disk units are removed at the same time, the validity of the data of the entire hard disk array 10 will be lost.

As described above, the present invention is not limited to such an embodiment, and can be implemented in various forms without departing from the gist of the present invention. For example, in the hard disk array system 1 of the above-described embodiment, the first LED 30a to 30d and the second LED 40 as the notification means are provided to the hard disk units 60a to 60d of A to D themselves.
a to 40d and the third LEDs 50a to 50d, the hard disk units 60a to 60d of A to D of the housing of the hard disk array 10 as shown in FIG.
The configuration may be provided at a position where the correspondence with is clear. However, in FIG. 3, the LEDs 130a to 130d correspond to the first LEDs 30a to 30d. LED140a ~
140d and LEDs 150a to 150d are also second L
It corresponds to the EDs 40a to 40d and the third LEDs 50a to 50d.

The LEDs for indicating a certain state are A to D.
In each of the hard disk units 60a to 60d, the LED for notifying another state may be provided at a position where the correspondence with the A to D hard disk units 60a to 60d of the housing of the hard disk array 10 becomes clear. . For example, the first and second LEDs for notifying the failure occurrence state and the data reconstruction state are provided in the respective hard disk units 60a to 60d of A to D, and the third LED for notifying the removable state is A of the housing of the hard disk array 10.
It is conceivable that the hard disk units 60a to 60d of D to D are provided at positions so as to be clearly associated with each other.

Further, in the above-mentioned embodiment, the LED as the notifying means is "fault occurrence state", "data reconstruction state",
Although the position is changed for each of the three states of "removable state", LEDs of different colors may be installed at the same position. For example, regarding the A hard disk unit 60a, the second L is located at the position of the first LED 30a shown in FIG.
ED40a, 3rd LED50a is also attached to the 1st LED
30a is green, 2nd LED40a is red, 3rd LED50a
May be configured so that the above three states can be identified by changing the color such as orange.
Instead of a liquid crystal display or the like, each of the A to D hard disk units 60a to 60d itself or the A to D hard disk unit 60a of the housing of the hard disk array 10 is used.
It may be provided at a position where the correspondence with ~ 60d becomes clear.

In the above embodiment, all three states of "fault occurrence state", "data reconstruction state" and "removable state" are detected and notified. It suffices as long as at least one of them is detected and notified. In addition, as a storage device array, R
Although the hard disk array 10 of AID4 is used, it is not limited to RAID4 as long as it is a storage device array in which at least one of the above three states can occur, and a semiconductor such as a silicon disk unit instead of the hard disk unit. The memory device may be used as a storage device.

[Brief description of drawings]

FIG. 1 is a block diagram of a hard disk array system.

FIG. 2 is an external view of a hard disk array.

FIG. 3 is an external view of a hard disk array according to another embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Hard disk array system 10 ... Hard disk array 20a-20d ... Hard disk 30a-30d ... 1st LED 40a-40d ... 2nd LED 50a-50d ... 3rd LED 60a ... A hard disk unit 60b ... B hard disk unit 60c ... C hard disk unit 60d ... D Hard disk unit 70 ... Main computer 71 ... Central control unit 73 ... Display 80 ... Controller 90 ... Input / output interface

Claims (2)

[Claims] 1. A storage device comprising a plurality of storage devices for storing data and at least one storage device for storing data protection information generated from the data, and when one of the storage devices fails. A storage device array capable of reconstructing the data stored in the failed storage device from the remaining storage devices, and further capable of removing one of the storage devices in the operating state, In a storage device array system including a main computer that controls the storage device array, the main computer has at least one of three states of a failure occurrence state, a data reconstruction state, and a removable state of each storage device. The state can be detected,
A storage device array system characterized in that the state of each of the detectable storage devices is notified to the outside by a notification means provided corresponding to each of the storage devices. 2. The storage device array system according to claim 1, wherein the notifying means is provided in each of the storage devices themselves.