JP2000339103A - Storage medium library array device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To execute a data restoration processing and a regular data reading/ writing processing in parallel and to improve operation efficiency by using at least one drive at every library unit as a restoration drive. SOLUTION: It is judged whether a media into which data is written at the time of degradation driving, namely, a restoration media exists or not in the media managed by a library array device (Si). When the restoration media exist, it is judged whether array constitution where the drive is idle, namely, all drives forming one group of array constitution mountable the restoration media and a degradation media are in idle states for executing a data restoration processing exists or not (S2). When such drives exist, 'mounting processing start instruction' and 'media restoration indication instruction' signals are transmitted (arrow X) to an array controller and the media restoration processing is executed (S14).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an arrangement of a plurality of portable storage media used as storage devices of a computer system and drive devices for reading / writing data from / to the storage media arranged in an array. In particular, the present invention aims to shorten the time for restoring the storage contents of a storage medium and improve the operation efficiency of the storage medium library array apparatus.

[0002]

2. Description of the Related Art In a computer system, there is a disk array device as one of external storage devices realizing a large capacity and high performance. 2. Description of the Related Art A conventionally known disk array device comprises a plurality of magnetic disk devices arranged in an array and operates each of the magnetic disk devices in parallel to read / read data divided and stored in each of the magnetic disk devices. This is a storage device that performs writing at high speed. In a disk array device, an exclusive OR (XO) of data on a plurality of magnetic disks (data disks) is performed.
By storing the calculation result (parity data) of R) in a magnetic disk (parity disk) device other than the plurality of magnetic disk (data disks) devices,
High reliability against data failure. That is, when certain data is lost (for example, when one of a plurality of magnetic disk devices fails),
The lost data can be restored using the data and parity data stored in the remaining magnetic disk devices. Such a disk array device is RA
The storage medium is a fixed magnetic disk (that is, a hard disk).

In such a disk array device, a large-capacity and relatively high-performance storage device can be realized by using a portable medium type storage device instead of the magnetic disk device. Examples of the portable medium type storage device include a magnetic tape device and an optical storage device. Particularly recently, DVD (Digital Versatile)
Disk). These storage devices are characterized in that a portable storage medium and a drive device for reading / writing data are separated from each other, and the storage medium is loaded into an arbitrary drive device to read / write data on the storage medium. That is. If the portable storage medium is inexpensive, a disk array system using the portable storage device can be realized at low cost.

In a large-scale computer system,
Libraries are introduced to facilitate the management of numerous storage media. The library includes, in addition to the storage medium and the drive device, a storage shelf (slot) for storing a large number of storage media, a transport holder for transferring the storage medium between the storage shelf and the drive device, and the like.
Since the data handled by computer systems is becoming increasingly large-scale, there is also a great need for improved reliability. Therefore, it is effective to realize high reliability and high performance by applying the disk array even in the storage device system configured by the portable medium as described above. As a technology in which a disk array is applied to a portable medium, RAIL (Redundant Arrays) is an array configuration in which a plurality of ordinary libraries composed of storage shelves, drive devices, transport holders, and the like are combined and arranged.
of Inexpensive Library
s) have been proposed (Alan E. Bell (IB)
M Research Division): DVD
Applications, COMDEX96, No
v. 20, 1996). In this case, in the related art, only one drive device is used in one library.

In a library array device using the portable medium type storage device, when the data and parity data to be divided and stored are divided and stored in n (n is an integer of 2 or more) storage media, When reading / writing data in the library, generally, the n storage media constituting a set of arrays stored in the storage shelf are taken out, and the transport holder is transported to the n drive devices. Then, the data is loaded into the n drive devices to read / write the data.

[0006]

By the way, in a conventionally known library array device using a portable medium type storage device, one of a plurality of drive devices does not operate at all or malfunctions. When an abnormality such as a defective storage medium mounted on the drive device occurs (that is, data cannot be read from or written to one of a plurality of storage media simultaneously processed in the data read / write processing). In the case), the degenerate operation is performed using a drive device of a system other than the system of the abnormal drive device to continue reading and writing of data. That is, n forming a set of arrays
In the (n-1) non-abnormal storage media among the storage media, degenerated data without redundancy is created. Then, the system administrator replaces the abnormal drive device (or storage medium) as a subsequent repair work, and replaces the plurality (n-1) of the drive devices that have been installed in the abnormal drive device after the completion of the degeneration operation. By reading data and parity data from the storage medium and performing a predetermined data restoration process, the contents of the data that should have been originally written to the predetermined storage medium by the failed drive device are restored,
This is written in the storage medium.

[0007] However, a great deal of time is required for the data restoration processing, and no other processing can be performed at all until the data restoration processing is completed. That is, there is a problem that the operation efficiency of the storage medium library array device is extremely low.

SUMMARY OF THE INVENTION The present invention has been made in view of the above points, and has been made in such a manner that data restoration processing can be performed simultaneously by a plurality of series of drives, and data restoration processing and normal data reading / writing processing can be performed in parallel. By doing so, it is intended to provide a storage medium library array device with extremely high operation efficiency.

[0009]

A storage medium library array device according to the present invention transports a portable storage medium between a storage unit and a drive, and writes or reads data to or from the storage medium in the drive. In a storage medium library array device which includes a plurality of units and drives the plurality of library units in parallel to read and write data in an array configuration, a plurality of drives are provided for each of the library units, Using at least one of the drives as a restore drive for each of the library units, a storage medium corresponding to at least one set of arrays to be restored is set in each restore drive, and data restoration processing is performed. It is characterized by comprising control means.

For example, a storage medium library array device has an array controller for mounting a storage medium on a drive of each library unit and controlling data read / write processing and the like. In this case, the control means is an array controller. Is included as one of the control functions. For example, an array controller including the function of the control unit automatically mounts a storage medium on a predetermined drive of each of the library units forming a set of array configurations not used during data restoration processing, and By controlling the array configuration in parallel, a new data restoration process can be performed without stopping each process performed for each array configuration in use.

According to this, when it is specified that the data restoration process is performed during the data read / write process or the like, an empty drive other than the one used in the data read / write process or the like is automatically selected. By automatically mounting and using a plurality of storage media required for data restoration processing on the empty drive, data read / write processing and data restoration processing (or multiple data restoration processing) can be performed in parallel. become able to. That is, the array controller mounts a medium on a drive that is not used among a plurality of drives arranged in the library unit. Then, data restoration processing is performed with the drive on which the medium is mounted as one set of array configuration. Since a plurality of drives are arranged in each storage system as described above, a plurality of array configurations can be configured. Therefore, different data read / write processing or data restoration processing can be performed in each array configuration. As described above, by automatically mounting the medium on a drive other than the drive used for the data read / write processing or the like, the data restoration processing can be performed without stopping the data read / write processing.

The present invention can be constructed and implemented not only as a device invention, but also as a method invention. Further, the present invention can be implemented in the form of a program of a computer or a processor such as a DSP, and can also be implemented in the form of a recording medium storing such a program.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a perspective view showing an example of the overall configuration of a storage medium library array device LA to which the present invention is applied. FIG. 2 shows the library units U1 to U in FIG.
6 is an enlarged perspective view showing an example of the entire configuration of FIG.
However, all of the library units U1 to U6 may have the same configuration. FIG. 2 shows an example of the entire configuration of only the library unit U1 alone. In the following description, the term “medium” simply includes not only the medium itself but also a tray on which the medium is mounted.
The storage medium library array device LA according to the present invention provides a control command such as a data read / write command or a data restoration command to the array controller AC from the operation panel P or a higher-level control device COM (for example, a server computer such as a personal computer). Are provided via a control interface (for example, SCSI interface or the like) (not shown) so that six library units (hereinafter simply referred to as units) U1 to U6 operate in parallel to read / read data from / to the medium. This is a device that performs writing at high speed. The array controller AC is composed of a microcomputer including an MPU, a ROM, a RAM, and the like (not shown), and controls the transport of the medium and the data read / write processing or the data restoration processing by driving a plurality of drives in parallel. The host device COM can also manage a plurality of library array devices LA.

In each of the units U1 (U2 to U6), based on a control command from the array controller AC, the control unit CT controls a large number of storage shelves Ta of the storage series T storing the medium (to simplify the description). (Only one is shown for this purpose), a desired medium is taken out from a predetermined storage shelf, and the transfer holder H is mounted so as to house (mount) the medium in one of a plurality of drives 10 to 13 provided. After the transport, the drives 10 to 13 are controlled to perform data read / write processing. That is, each unit U1 (U2 to U
6) is configured so that each unit U1 (U2 to U6) alone can perform data read / write processing on a medium. Further, by operating a plurality of drives 10 to 13 of each unit U1 (U2 to U6) in parallel, different data read / write processing can be performed. Further, each unit U1 (U2 to U6) has a mass entry M1 (M2 to M6).
6) is provided so that the medium can be introduced into the unit from the outside or discharged from the unit to the outside.

The above-described storage medium library array device LA
Then, at the time of data read / write processing, a medium is mounted on one of a plurality of drives arranged in the storage system.
If the drive on which the medium is mounted is not usable in each storage system, the drive on which the medium is mounted is replaced with one.
Data read / write processing is performed as a set of array configurations. Since a plurality of drives 10 to 13 are arranged in each storage system U1 (U2 to U6) as described above, a plurality of array configurations can be configured. Therefore, different data read / write processing can be performed in each array configuration. In addition,
The storage medium library array device LA in the present embodiment is not limited to this. For example, the units U1
The number of U6 is not limited to six, and a plurality of units may be provided, and at least one unit is provided in the apparatus and is shared by the units U1 to U6 without providing the transport holder H for each unit U1 to U6. Alternatively, the medium may be transported. Furthermore, 4 for each unit U1 to U6
Although the example in which the drives 10 to 13 are provided has been described, the present invention is not limited to this, and the drives 10 to 13 may have a plurality of drives for each of the units U1 to U6.

The above-mentioned storage medium library array device L
In A, when performing a data write process, one of a plurality of drives forming a set of array configurations on which a medium is mounted does not operate at all or malfunctions, or is mounted on a plurality of drives. If an abnormality such as one of the defective media is defective,
As in the case of the conventional storage medium library array device, the data writing process is continued using a drive of a series other than the series of the drive. That is, the degenerate operation is performed. After the end of the degeneration operation, it is necessary to write data to a medium on which data could not be written (that is, a medium loaded in a drive having an error or a medium having an error). Therefore, data (or parity data) written to a plurality of media loaded in a drive in which no abnormality has occurred is read, and a data restoration process described later is performed to restore a medium to which data could not be written. Write the written data. In the storage medium library array device LA according to the present invention, each unit U1
Since a plurality of drives 10 to 13 are provided for each of the drive units U6 to U6, data recovery can be performed quickly by using these drives efficiently. It should be noted that the drive to be used in the data restoration process is determined by the host device C
There are two patterns, one determined on the OM side and one determined on the array controller AC side.

FIG. 3 shows that the host device C
OM and array controller AC (the upper right is the upper device,
FIG. 3 is a control flow diagram for conceptually explaining each process performed by an array controller (on the left side). This series of data restoration processing is started from processing on the host device side when a “data restoration processing command” is given to the host device COM. First, it is determined whether or not there is a restoration medium among the media (storage media) managed by the library array device LA (step S1). That is, when one of the drives forming the array configuration is unusable during normal data writing, a degenerate operation is performed in the data writing process, and the unusable operation is performed. Data is written only to media mounted on drives other than the drive. In this case, the higher-level device manages a medium on which data is written during the degenerate operation (that is, a medium mounted on a drive other than an unusable drive) as a restoration medium. Conversely, the higher-level device manages a medium on which data cannot be written during the degenerate operation (that is, a medium mounted on an unusable drive) as a degenerate medium. For a degraded medium to which data could not be written, data can be restored and written from a restoration medium. Therefore, it is determined whether or not there is a medium for which data restoration processing needs to be performed among the media managed by the library array device. As described above, in this embodiment, the necessity of the data restoration processing is grasped and managed by the host device COM.

If there is no restoration medium among the media managed by the library array device LA (NO in step S1), the process ends. That is, since there is no degraded medium and there is no need to perform data restoration processing on all media managed by the library array device, the processing ends. If there is a restoration medium (YES in step S1), it is determined whether or not the drive has free space (step S1).
2). That is, whether or not there is an array configuration in which all the drives forming a set of array configurations on which the recovery media and the degraded media can be mounted in order to perform the data recovery process is empty (Whether or not there is a set of array configurations that do not exist). If there is no such drive, that is, if all array configurations are being used for other data read / write processing or data recovery processing, recovery media and degraded media cannot be mounted on the drive, and therefore Since new data restoration processing cannot be performed, the processing ends (NO in step S2). On the other hand, if there is such a drive, that is, if there is an unused array configuration in the array configuration (Y in step S2)
ES), the host apparatus COM transmits a "mount processing start command" and a "media restoration instruction command" signal to the array controller AC (see arrow X), and the host apparatus CO
The processing on the M side ends. As described above, in this embodiment, the individual drives 10 to 13 in the units U1 to U6 are used.
Is used and managed by the host device COM.

The array controller AC includes a host device CO
When the “mount processing start command” is received from M, the following processing is started. First, "Mount processing start instruction"
Upon receiving the signal, the “normal mounting process” is performed, and the restoration medium and the degenerated medium are mounted on the drives forming a set of array configurations determined to have free space (step S11). If the command received from the higher-level device does not include the “media restoration instruction command” (for example, during normal data read / write processing) (NO in step S12), the process ends. If the command received from the higher-level device includes a “media restoration instruction command” (YES in step S12), it is determined whether the medium mounted on each drive is a restoration medium or a degenerated medium (step S13). ). Data restoration is performed by reading data from a restoration medium and writing the restored data to a degraded medium. Therefore, it is determined whether the medium mounted on the drive is a restoration medium or a degraded medium.
That is, if the media has been manually replaced by the user and the media information managed by the library array device does not match the information of the actually stored media, data cannot be restored. Therefore, the media is checked for that. If the medium is not a restored medium or a degenerated medium (NO in step S13), the process ends without performing data restoration. If the medium is a restoration medium or a degenerated medium (YES in step S13), a medium restoration process is performed (step S14). That is, a process of reading data from the restoration medium, reconstructing the data, and writing the reconstructed data to the degraded medium is performed.

In the above processing on the higher-level device COM,
Whether the data restoration process is performed or not is automatically determined based on whether or not there is a restoration medium in the media managed by the library array device LA (see step S1). May be directly instructed to the host device to perform data restoration. In this case, the user can designate a restoration medium or a degraded medium necessary for restoring data or a drive used for restoring data. Alternatively, when the user specifies normal data read / write processing and the specified medium is a restoration medium or a degenerated medium, the data read / write processing may be performed after the data restoration processing is automatically performed. .

Here, the above-described data restoration processing will be specifically described with reference to FIG. FIG. 4 is a schematic front view schematically showing only the units of the storage medium library array device from the front. However, all of the units U1 to U6 are illustrated side by side in order to facilitate understanding of the description. It should be noted that the numbers attached to the right end of the figure indicate drive mount rows, and a drive having the same drive mount row forms one set of array configuration.
The mounting of the medium on each drive is performed by designating the drive mount line. In this embodiment, for convenience,
Drive mount rows corresponding to the respective drives are indicated by 0 to 3 in order from the top. In the following description, the drive of “mount 0” of the unit U1 is “drive 10”, and the drive of “mount 3” of the unit U6 is “drive 63”.
The drive of each unit U1 to U6 is shown as follows.

For example, if data read / write processing has already been performed on the drives 10 to 60 forming a set of array configurations corresponding to “mount 0” (refer to the first line O), the data It is assumed that a "restore instruction command" has been given. Then, since the drive of “mount 0” is in use by the host device, the other “mount 1” to
Drives 11 to 11 in an array configuration corresponding to “mount 3”
It is determined whether any of the drives 61 (12 to 62, 13 to 63) is not in use. Now, "Mount 0"
Assuming that all the drives 11 to 61 (12 to 62, 13 to 63) other than the drives 10 to 60 having the array configuration corresponding to the “mount 1” are empty, the drives 11 to 61 having the array configuration corresponding to “mount 1” are selected. Then, the array controller mounts the media on the drives 11 to 61 (see the second line P). Thus, drives 11 to 6
1. If the mounted media are restoration media (see the shaded ellipse) or degenerated media (see the filled ellipse), data restoration is performed. Next, normal data read / write processing is performed in this state, that is, in the array drive 10 to 60 corresponding to “mount 0” (refer to the first line O), and the array drive corresponding to “mount 1” is performed. In a state where data restoration processing is being performed in steps 11 to 61 (see the second line P), when a new normal “data read / write instruction command” is given, a drive having an array configuration corresponding to “mount 2” Data read / write processing is performed in 12 to 62 (see the third line Q). Further, when a "data restoration instruction command" is given, data restoration processing is performed in the drives 13 to 63 in the array configuration corresponding to "mount 3" (see the fourth line R).

As described above, in the library array device, each process can be performed in parallel using a plurality of drives, and in the data restoration process, an available drive is automatically selected to select a medium. By mounting, data restoration can be performed in parallel even during data read / write processing or data restoration processing. In the above-described embodiment, a plurality of array configurations (see lines O to R) can be made by assembling six drives, and each line O to
R can perform different data read / write processing or data restoration processing in parallel. That is, data restoration processing can be performed without stopping other data read / write processing, and a plurality of data restoration processing can be performed in parallel. In the above description, an empty drive is searched for from "mount 0" to "mount 3". However, the present invention is not limited to this, and it may be possible to search for "mount 0" from "mount 3". Alternatively, a random search may be performed.

Further, at least one drive for data restoration may be provided as a spare drive for each unit as a spare drive. FIG. 5 shows an embodiment in which the drives 13 to 63 in the array configuration indicated by “mount 3” are arranged as spare drives for each of the units U1 to U6. In FIG. 5, "mount 0" to "mount 2"
Drives 10 to 60 (11 to 6) having an array configuration corresponding to
1, 12-62) are used for normal data read / write processing, and drives 1 in an array configuration corresponding to "mount 3"
3 to 63 are used as spare drives. In this embodiment, at the time of data restoration processing, restoration media or degenerated media are accommodated in the spare drives 13 to 63. That is, when the "data restoration instruction command" is designated, the restoration medium or the degraded medium is automatically mounted on the spare drives 13 to 63 to perform the data restoration (see the third line N). By doing so, the data restoration process can be performed simultaneously with the normal data read / write process. In the above-described embodiment, only one set of drives forming an array configuration is used as a spare drive. However, the present invention is not limited to this, and a plurality of spare drives are configured, and data restoration processing is performed in parallel with a plurality of spare drives. Needless to say, the configuration may be such that the operation can be performed.

Next, a case where the necessity of data restoration processing is grasped and managed on the array controller AC side, which is different from the above-described embodiment, will be described. FIG. 6 is a flowchart for conceptually explaining the processing performed by the array controller AC. This series of processing is started by receiving a predetermined signal from the host device COM. First, the library array device LA is connected to the host device C.
It is determined whether or not the signal received from the OM is a "data read / write instruction command" (step S21). If the received signal is not the “data read / write instruction command” (NO in step S21), the process jumps to step S23. If the received signal is a "data read / write instruction command" (YES in step S21), the flag F1 is set (step S22). Next, it is determined whether data restoration is necessary (step S23). If data restoration is not required (NO in step S23), the process jumps to step S25. If data restoration is necessary (step S2
3; YES), a flag F2 is set (step S2).
4). That is, when the "data read / write instruction command" is received, the flag F2 is automatically set if the medium to be read / written is a restoration medium or a degenerated medium. On the other hand, the operation panel P
, The flag F2 is set unconditionally. Then, it is determined whether the flag F1 or the flag F2 is set (step S25). Flag F1 and flag F
If both are not set (step S25)
NO), the process ends. If both or one of the flags F1 and F2 is set (YES in step S25), step S26 is performed.
Is performed. In step S26, when only the flag F1 is set, a predetermined normal data read / write subroutine process is started, and when only the flag F2 is set, a predetermined data restoration subroutine process is started. When both the flag F1 and the flag F2 are set, the data restoration subroutine process is started, and the normal data read / write subroutine process is started after the completion of the data restoration subroutine process. Needless to say, each of these processes is performed using a drive that is not currently used (that is, is not used). Thus, in this embodiment, each unit U1
The use status of the individual drives 10 to 13 in U6 to U6 is grasped and managed by the array controller AC.

It is to be noted that how many drives among the plurality of drives 10 to 13 of each of the units U1 to U6 are to be used for data restoration is set in the operation panel P in advance, and the user selects the setting so that the data can be selected. The number of drives to be restored may be specified.
In other words, the drive dedicated to data read / write and the drive dedicated to data restoration can be used by dividing them into an arbitrary number, so that the drives can be used efficiently according to the usage mode. The use efficiency of the library array device AL is improved. For example, if there are many storage media for which data is to be restored, the data restoration process may be performed on all drives.If there is no storage medium for which data is to be restored, all drives are used for data read / write processing. I just need. In this way, each process can be completed in a short time. Further, in the storage medium library array device LA, each unit U1 to U6
May be managed, and the drives to be used may be allocated based on the frequency of use. The management of the use frequency may be configured to be performed on the host device COM side, or may be configured to be performed on the array controller AC side. For example, when the usage frequency is managed on the array controller AC side, the usage frequency of the drive specified by the higher-level device COM is compared with the usage frequency of other drives in the same unit as the specified drive. A drive with a low usage frequency may be automatically allocated and used by the array controller AC. This makes it possible to average the frequency of use of the drives 10 to 13 in each of the units U1 to U6, thereby reducing drive failures. That is, the number of times of the degeneration operation can be reduced.

[0028]

As described above, in the present invention, a plurality of data restoration processes can be performed in parallel, so that a large number of data restoration processes can be performed at once, and therefore the time required for data restoration can be reduced. Will be able to In addition, since the data restoration process can be performed without stopping the data read / write process, the storage medium library array device can be used efficiently, and thus the processing efficiency can be improved.

[Brief description of the drawings]

FIG. 1 is a perspective view showing an overall configuration example of a storage medium library array device to which the present invention is applied.

FIG. 2 shows library units U1 to U6 in FIG.
FIG. 2 is an enlarged perspective view showing an example of the entire configuration of FIG.

FIG. 3 is a control flowchart for conceptually explaining each process performed by a host device and an array controller during a data restoration process.

4 is a diagram for explaining the data restoration process shown in FIG. 3, and is a schematic front view schematically showing the unit configuration of the storage medium library array device from the front.

FIG. 5 is an explanatory diagram for describing data restoration processing when a spare drive is configured for each unit of the storage medium library array device.

FIG. 6 is a flowchart illustrating an example of a process performed by an array controller.

[Explanation of symbols]

AC: Array controller, P: Panel, U1 to U6 ...
Library unit, M1 to M6 ... mass entry, T ...
Storage system, Ta: storage shelf, CT: control unit, H: transport holder, 10 to 60 (11 to 61, 12 to 62, 13 to 6)
3) Drive, AL Storage medium library array device, COM Host controller

Claims (6)

[Claims] 1. A portable storage medium is transported between a storage unit and a drive, and a plurality of library units for writing or reading data to or from the storage medium in the drive are provided. A storage medium library array device that drives and reads / writes data in an array configuration, comprising a plurality of drives for each of the library units, wherein at least one of the drives is restored for each of the library units during data restoration processing. A storage medium library array device comprising a control unit for setting a storage medium corresponding to at least one set of arrays to be restored to each drive for use as a drive and performing data restoration processing. . 2. The storage medium library array device reads and writes array configuration data in accordance with an instruction from a higher-level device, and uses the other one drive of each library unit as a normal data read / write drive. 2. The storage medium library array device according to claim 1, wherein the data restoration process can be performed simultaneously and in parallel while performing normal read / write processing of array configuration data. 3. 3. The storage medium library array device according to claim 2, wherein the data restoration process is executed in accordance with an instruction from the host device. 4. The storage medium library array device according to claim 2, wherein said data restoration processing is executed in accordance with control or manual instruction in said storage medium library array device. 5. In a data restoration process, a plurality of drives are used as restoration drives for each library unit, and storage media corresponding to a plurality of sets of arrays to be restored are set in each restoration drive. 5. The storage medium library array device according to claim 1, wherein data restoration processing for a plurality of sets of arrays can be performed simultaneously and in parallel. 6. The storage medium library array device according to claim 5, further comprising means for selecting a mode of how many sets of arrays in one or more sets are to be subjected to data restoration processing simultaneously and in parallel.