JP2000122811A - Storage device managing method and storage device management system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent data arranged in an ECC generation unit from being entirely eliminated when a failure takes place at the time of receiving write data from an information processor, reading insufficient data in the ECC generation unit from a storage device, generating an ECC after arranging the data in the ECC generation unit and writing the data and the ECC in the storage device. SOLUTION: When a failure takes place (4020) when write data (4904) that are smaller than an ECC generation unit are received (4002) from a host computer (10), insufficient data in the ECC generation unit are read (4005) from a DVD-RAM drive (30), an ECC is generated after the data are arranged in the ECC generation unit and the data and the ECC are written in the DVD-RAM drive (30), the data arranged in the ECC generation unit are saved in a flash memory (25). Thus, not only the write data that are smaller than the ECC generation unit but also the entire ECC block data are prevented from being eliminated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a storage device management method and a storage device management system.
The write data smaller than the redundant data generation unit is received from the information processing device, the data lacking in the redundant data generation unit is read from the storage device, the data is aligned in the redundant data generation unit, and then the redundant data is generated. The present invention relates to a storage device management method and a storage device management system capable of guaranteeing that, even when a failure occurs when writing data and redundant data, all data aligned in the redundant data generation unit is not lost.

[0002]

2. Description of the Related Art Recently, as a storage device connected to an information processing apparatus (host computer), attention has been paid to:
DVD-RAM (Digital Versatile Disc-Random Acc
ess Memeory) device. For details on DVD-RAM devices, see Toshiba Review 2, 1998, VOL. 53 NO.
I: The expanding DVD world "and the National Technical Repo
rt, 1997 JUNE, "Special DVD" etc.

FIG. 19 shows a data management method in the DVD-RAM drive 30. DVD-RAM drive 30
In, data management is performed in units of 2 KB blocks. On the other hand, ECC (Error Check Code) management is performed in units of 16 blocks (referred to as ECC blocks) in order to enhance the data regeneration capability.

[0004] As shown in FIG. 20, the host computer reads and writes data from and to the DVD-RAM device 31 in units of 2 KB blocks. DVD-RAM device 3
In No. 1, reading from the DVD-RAM drive 30 is performed in block units, but writing to the DVD-RAM drive 30 is performed in ECC block units. For example, FIG.
0, when the host computer requests writing of the data A of the block 4, the DVD-RAM device 3
1 receives the data A and stores it in the buffer 29004 (20001). Next, the DVD-RAM device 31
Data of ECC block 0 (block 0 to block 15) including block 4 is stored in DVD-RAM drive 3
Read from 0 to the buffer 29500 (20005),
After merging the data A into the block 4 in the buffer 29500, and then generating the ECC (2001)
0), the data of the ECC block 0 in the buffer 29500 and the generated ECC are stored in the DVD-RAM drive 30.
And write it to a DVD-RAM medium (2002
0).

[0005]

A failure (power failure, instantaneous power, laser emission stop, failure of DVD-RAM medium, failure of DVD-RAM drive 30, etc.) occurs during writing of ECC block 0 shown in FIG. Then, since the data of the ECC block 0, that is, the data of the blocks 0 to 15 and the ECC do not match, the writing of all the data of the blocks 0 to 15 has failed. That is, the conventional DVD-RAM device 31
In this case, when writing of data of one block fails, there is a problem that all data of the ECC block including the block is lost. Then, the host computer 10 can recognize from the failure report that the writing of the data of the block requested to be written has failed, but cannot recognize which data other than the data has been lost, which may cause a greater failure. There was a problem.

Accordingly, an object of the present invention is to receive write data smaller than a redundant data generation unit from an information processing device, read data lacking in the redundant data generation unit from a storage device, and align the data in the redundant data generation unit. Storage device management method and storage device capable of guaranteeing that even when a failure occurs when redundant data is generated and the data and the redundant data are written to the storage device, all data aligned in the redundant data generation unit is not lost. To provide a management system.

[0007]

According to a first aspect, the present invention provides a method for receiving write data smaller than a redundant data generation unit from an information processing apparatus, reading data lacking in the redundant data generation unit from a storage device, After aligning the data in the generation unit, redundant data is generated, and the data and the redundant data are written in the storage device. If the writing fails, the data aligned in the redundant data generation unit is stored in a nonvolatile storage. Means, and when there is a read request from the information processing apparatus for data belonging to the data aligned with the redundant data generation unit, the data held in the nonvolatile storage means is used preferentially, and Provides a storage device management method characterized by not reading. In the above configuration, the redundant data is, for example, ECC or RAID (Redundant Arrays of
Parity in Inexpensive Disks). In the storage device management method according to the first aspect, when writing fails, the data aligned with the redundant data generation unit is held in a non-volatile storage unit, and when a read request is issued from the information processing device, the non-volatile storage unit stores the data. The data held in the sex storage means is used preferentially. For this reason, it is possible to guarantee that all the data aligned in the redundant data generation unit will not be lost.

In a second aspect, the present invention provides a write data receiving means for receiving write data smaller than a redundant data generation unit from an information processing apparatus, and a data read means for reading data insufficient for a redundant data generation unit from the storage device. Data writing means for writing redundant data after the data is aligned in the redundant data generation unit and writing the data and the redundant data to the storage device; and aligning the data in the redundant data generation unit when the writing fails. Non-volatile storage means for holding data, and data for preferentially using the data held in the non-volatile storage means when a request for reading data belonging to the data of the redundant data generation unit is received from the information processing device. A storage device management system comprising a priority use unit is provided.
In the storage device management system according to the second aspect, the first
The storage device management method according to the above aspect can be suitably implemented. Therefore, it is possible to guarantee that all the data aligned in the redundant data generation unit will not be lost.

In a third aspect, the present invention is characterized in that the non-volatile storage means is a non-volatile memory or a battery-equipped memory detachable from a storage device management system. Storage device management system is provided. In the storage device management system according to the third aspect, the nonvolatile memory or the battery-equipped memory is removed from the storage device management system, and the data is transferred to another storage device management system, thereby recovering data in the other storage device management system. Can be.

According to a fourth aspect of the present invention, in the storage device management system having the above-mentioned configuration, when the normal power supply to the storage device management system is stopped, the data aligned in the write data management unit is stored in the nonvolatile storage device. There is provided a storage device management system characterized by including an emergency power supply device for backing up power to the storage device management system at least for a time necessary for saving to the storage means. In the storage device management system according to the fourth aspect, when the normal power supply to the storage device management system is stopped,
The data aligned in the redundant data generation unit can be saved in the nonvolatile storage unit. Therefore, even when the normal power supply to the storage device management system is stopped, it is possible to guarantee that all the data aligned in the redundant data generation unit will not be lost.

The storage device management system of the present invention may constitute a RAID in combination with a plurality of storage devices. Further, the storage device management system of the present invention may be provided between the information processing device and the storage device, or may be provided inside either the information processing device or the storage device.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to the drawings. Note that the present invention is not limited by this.

<< First Embodiment >> FIG. 1 shows a first embodiment of the present invention.
1 shows a storage device management system according to an embodiment. The storage device management system 20 includes a host computer 10
And a DVD-RAM drive 30. The connection between the storage device management system 20 and the host computer 10 is made via the fiber channel 15, and the connection between the storage device management system 20 and the DVD-RAM drive 30 is made using SCSI.
(Small Computer System Interface) 25.

FIG. 2 shows a system configuration of the storage device management system 20. The storage device management system 20 includes a fiber channel interface 210 and an S
A CSI interface 220, which are PCI
(Peripheral Component Interconnect) bus 290. A CPU 280 and a main memory 284 are connected to the PCI bus 290 via a chipset 286 for controlling the flow of programs and data, and a secondary cache 282 is connected to the CPU 280. Further, the PCI bus 290 has a PCMCIA
(Personal Computer Memory Card International Asso
ciation) The flash memory 250 is connected via a controller 259 and a PCMCIA connector 255.

The main memory 284 stores a control program and the fiber channel interface 210
And temporarily stores the request and data received from the host computer 10 via the SCSI interface 220 and the data received from the DVD-RAM drive 30 via the SCSI interface 220.

The normal data flow will be described with reference to FIG. When the host computer 10 requests data writing for one block, the storage device management system 20 receives the data of that block and sends it to the main storage 2.
It is held in the segment 3904 in the block 84 (3002).
The CPU 280 reads the ECC block including the block from the DVD-RAM drive 30 and
4 (3005). Next, the data of the segment 3904 is combined with the read data of the ECC block (3007) to generate an ECC. Then, the data of the ECC block and the ECC are written to the DVD-RAM drive 30 (3009).

Referring to FIG. 4, a DVD-RAM drive 3
The data flow when a failure occurs when writing data to 0 will be described. When the host computer 10 requests data writing for one block, the storage device management system 20 receives the data of the block and stores it in the segment 4904 in the main memory 284 (4002). The CPU 280 reads the ECC block including the block from the DVD-RAM drive 30 to the main memory 284 and holds the ECC block in the main memory 284 (400
5). Next, the data of the segment 4904 is combined with the read data of the ECC block (4007),
Generate Then, the data of the ECC block and the ECC
Write to the DVD-RAM drive 30 (4
009). At this time, if a failure occurs in the DVD-RAM drive 30 (4020) and the data cannot be written, the data of the ECC block in the main memory 284 is written to the flash memory 250 (4010).
Thereafter, even when a data write request is issued from the host computer 10, the storage device management system 20 refuses to receive the data. When the host computer 10 requests the reading of the data of the block belonging to the ECC block, if the corresponding data exists in the main memory 284, the data is transferred and the corresponding data is stored in the main memory 284. Is not present, the data held in the flash memory 250 is transferred and the DVD
-Data is not read from the RAM drive 30.

Referring to FIG. 5, a DVD-RAM drive 3
The data flow when a failure occurs when reading data from 0 will be described. When the host computer 10 requests data writing for one block, the storage device management system 20 receives the data of the block and holds it in the segment 5904 in the main memory 284 (5002). The CPU 280 attempts to read an ECC block including the above-mentioned block from the DVD-RAM drive 30 to the main memory 284 (5005). At this time, D
A failure occurs in the VD-RAM drive 30 (500
6) If the data of the ECC block cannot be read, the data of the segment 5904 in the main memory 284 is written to the flash memory 250 (5010).
Thereafter, even when a data write request is issued from the host computer 10, the storage device management system 20 refuses to receive the data. When the host computer 10 requests to read the data of the block, if the corresponding data exists in the main memory 284, the data is transferred, and if the corresponding data does not exist in the main memory 284, Transfers the data held in the flash memory 250, and does not read data from the DVD-RAM drive 30.

The failure that has occurred in the DVD-RAM drive 30 is a failure other than that of the DVD-RAM medium,
If there is no failure in the AM medium, the failed DVD-
The DVD-RAM medium in the RAM drive 30 is taken out, and the D-RAM connected to another storage device management system (20) is taken out.
Insert it into the VD-RAM drive (30). Further, the storage device management system 20 connected to the failed DVD-RAM drive 30 sends the flash memory 25
0, and connect the flash memory 250 to the storage device management system (20) in which the DVD-RAM medium is inserted. When the storage device management system (20) is started in this state, the CPU (280) determines whether or not data is held in the flash memory 250. If the data is held, the CPU (280) stores the data in the DVD-RAM. Write to media. That is, E containing the block of data
Transfer the data of the CC block to the DVD-RAM drive (3
0), combines the data in the flash memory 250 to generate an ECC, and writes the data of the ECC block and the ECC to the DVD-RAM drive (30). Thus, the flash memory 250 and the DVD-R
The AM medium and the other storage device management system (20) and the DV
If you move the data to the D-RAM drive (30),
Can be recovered on VD-RAM media.

In the first embodiment, a buffer for two ECC groups (for receiving data from the host computer 10 and reading and combining data from the DVD-RAM drive 30) is used in the main memory 284. ,
By providing a plurality of areas in the main memory 284, LRU (Least
DVD-RAM drive 3 by managing data using cache algorithms such as Recently Used.
The number of times of reading and writing to 0 may be reduced so that a processing request from the host computer 10 can be responded at a higher speed. When a plurality of areas are provided, a large amount of data that has not been written to the DVD-RAM drive 30 exists in the main memory 284. Therefore, it is desirable to use the flash memory 250 having a sufficient capacity to save the data.

When the flash memory 250 is used,
When the entire system including the storage device management system 20 suffers a failure such as a power failure, data cannot be saved to the flash memory 250, and data may be lost. Therefore, as shown in FIG.
An atic random access memory (251) may be used instead of the flash memory 250 of FIG. in this case,
The buffer is provided not in the main memory 284 but in the SRAM 251 with a battery. As a result, even when a failure such as a power failure occurs in the entire system, the data can be retained. When the storage device management system 20 starts up after recovery from a failure or the like, the CPU 280 executes the SR with battery.
It is determined whether data is stored in the AM 251 and if so, the data is stored in the DVD-RAM.
Write to media.

Further, as shown in FIG. 22, power may be supplied from the non-stop power supply 299 to the storage device management system 20. Thus, even when a failure such as a power failure occurs in the entire system, the data in the main memory 284 can be saved to the flash memory 250.

According to the first embodiment, the DVD-
Even when data writing to the RAM drive 30 fails, all data in the ECC block is stored in the flash memory 250 or the SRAM 251 with battery.
Loss of all data in the C block can be prevented.

<< Second Embodiment >> FIG. 6 shows a second embodiment of the present invention.
1 shows a storage device management system according to an embodiment. This storage device management system 21
Is incorporated in the changer device 40 connected to the control unit. The changer device 40 includes the storage device management system 21.
And a plurality of DVD-RAM drives 30 and a DVD-RA
A plurality of hangars 420 holding M media, and a DVD-RA
A media transport unit 450 for transporting the M medium and a robot control circuit 410 for controlling the media transport unit 450 are provided.

The host computer 10 and the storage device management system 21 are connected by the fiber channel 15, and the storage device management system 21 and the DVD-RAM drive 30
The storage device management system 21 and the robot control circuit 410 are connected by the CSI 25, and the RS232C interface 4
11 is connected.

When the storage device management system 21 instructs the robot control circuit 410 to transport the DVD-RAM medium,
The robot control circuit 410 reads the DVD-
The media transport unit 4 transfers the RAM medium to the DVD-RAM drive 30 or returns the DVD-RAM medium in the DVD-RAM drive 30 to the storage 420.
50 is controlled.

FIG. 7 shows a system configuration of the storage device management system 21. The storage device management system 21 has basically the same configuration as the storage device management system 20 shown in the first embodiment, and the operation of writing data and holding data in the event of a failure will be described in the first embodiment. As you did. In addition, the CPU 280 receives the request for transporting the DVD-RAM medium from the host computer 10, instructs the robot control circuit 410 via the robot control interface 240, and executes the transport of the DVD-RAM medium.

In the second embodiment, as in the first embodiment, a cache algorithm may be used, an SRAM with a battery may be used, or a non-stop power supply may be used.

According to the second embodiment, in addition to the effects of the first embodiment, the storage device management system 21 is realized as a part of the changer device 40, thereby realizing the implementation of the storage device management system 21. Cost can be reduced.

<< Third Embodiment >> FIG. 8 shows a third embodiment of the present invention.
1 shows a storage device management system according to an embodiment. The storage device management system 22 includes the host computer 10
And four changer devices 80. The host computer 10 and the storage device management system 22 are connected by the fiber channel 15, and the storage device management system 2
2 and the changer device 80 are connected by SCSI25.

FIG. 9 shows the structure of the changer device 80. The changer device 80 includes a plurality of storages 420 for holding DVD-RAM media, a media transport unit 450 for transporting DVD-RAM media between the storages 420 and the DVD-RAM drive 30, and a media transport unit 450 for the media. Robot control circuit 410 to control
And a changer SCSI interface 40 for receiving a DVD-RAM medium transport request from the storage device management system 22 and converting it into a command to the robot control circuit 410.
5 is provided. DVD- in the changer device 80
The RAM drive 30 is connected to the storage device management system 22 via SCSI 25.

FIG. 10 shows a system configuration of the storage device management system 22. This storage device management system 22
The configuration is basically the same as that of the storage device management system 20 shown in the first embodiment, but in addition, four SCSI interfaces 220 for connecting to the four changer devices 80 are provided. Thus, each DVD-RAM drive 3 in the four changer devices 80
0 is combined to realize a RAID configuration. That is, three of the four DVD-RAM drives 30 are data storage drives, and one is a parity storage drive in RAID. In addition, the CPU 280 sets the RAI
In order to realize the D processing, a data distribution module for distributing data to three data storage drives, a parity generation module for generating parity from data distributed to the three data storage drives, and a read request from the host computer 10 When the data is read, the data read from the three data storage drives is collected and the host computer 1
A data collection module for transferring data to the data storage module 0 and a data recovery module for recovering data of the failed data storage drive from the data of the remaining two data storage drives and the parity of the parity storage drive when one data storage drive fails. And have implemented.

The normal data flow will be described with reference to FIG. The host computer 10 writes data to one logical block in the storage device management system 2
2, the data distribution module of the CPU 280 calculates the DVD-RAM drive corresponding to the data and the physical block in the DVD-RAM drive, receives the data from the host computer 10, and stores the data in the segment 11936 in the main memory 284. Hold (11
002). Next, the CPU 280 converts the data of the ECC block including the physical block into a D corresponding to the data.
Read from VD-RAM drive 30 (11023)
At the same time, data of a RAID group including the ECC block (data of a group of ECC blocks used to generate one parity) is read from another DVD-RAM drive 30 to the main storage 284 (11021, 11).
022). Next, the data of the segment 11936 is combined with the data read from the corresponding DVD-RAM drive 30 (11070). Then, an ECC is generated and set as a new ECC block. Next, the CPU 280
Generates a parity using the new ECC block and other data. Next, the data of the new ECC block is written to the corresponding data storage drive (11083), and the parity is written to the parity storage drive (11084).

Referring to FIGS. 12 and 13, one DVD-
The flow of data when a failure occurs when writing data to the RAM drive 30 will be described. In FIG. 12, when the host computer 10 requests the storage device management system 22 to write data to one logical block, the data distribution module of the CPU 280 sets the DVD-RAM drive corresponding to the data and the DV
A physical block in the D-RAM drive is calculated, data is received from the host computer 10, and the
84 in segment 12936 (1200
2). Next, the CPU 280 converts the data of the ECC block including the physical block into a DVD corresponding to the data.
-Read from the RAM drive 30 (12023) and, at the same time, read the data of the RAID group including the ECC block from another DVD-RAM drive 30 into the main memory 284.
(12021, 12022). Next, the data of the segment 12936 is combined with the data read from the corresponding DVD-RAM drive 30 (1207).
0). Then, an ECC is generated and set as a new ECC block. Next, the parity generation module of the CPU 280 generates a parity using the new ECC block and other data. Next, the data of the new ECC block is written to the corresponding data storage drive (12083), and the parity is written to the parity storage drive (12084). At this time, it is assumed that a failure has occurred in the data storage drive to which data is to be written (12006), and the data cannot be written.

Then, as shown in FIG. 13, the data of the new ECC block that could not be written is written to the failure data holding areas of all the other DVD-RAM drives 30 (13011, 13012, 1301).
4). These fault data holding areas are special areas provided separately from areas for writing normal data or parity. Thereafter, even if the host computer 10 requests data writing, the storage device management system 22
Refuse to receive data. Also, a new EC that could not be written from the host computer 10
When reading the data of the C block is requested,
If the data exists in the main memory 284, the data is transferred. If the corresponding data does not exist in the main memory 284, the data stored in the fault data holding area is read and transferred, and a fault occurs. It does not read data from the data storage drive.

The data flow when all the DVD-RAM drives 30 cannot be used will be described with reference to FIG. When all the DVD-RAM drives 30 become unusable, writing of data of a new ECC block that could not be written to the failure data holding area also fails (14011, 14012, 1401).
4). Therefore, the CPU 280 saves the data of the new ECC block that could not be written to the flash memory 250 (14050). Thereafter, even if the host computer 10 requests data writing,
The storage device management system 22 refuses to receive the data. When the host computer 10 requests to read data of a new ECC block that could not be written, if the data exists in the main memory 284, the data is transferred. If there is no data corresponding to, the data saved in the flash memory 250 is read and transferred, and the data is not read from the failed data storage drive.

In the third embodiment, as in the first embodiment, a cache algorithm may be used together, an SRAM with a battery may be used, or a non-stop power supply may be used.

According to the third embodiment, the RAID
Even when the configuration is adopted, the effects of the first embodiment can be obtained.

<< Fourth Embodiment >> FIG. 15 shows a storage device management system according to a fourth embodiment of the present invention. This storage device management system includes a DVD-RAM drive 3
0 is implemented as a storage device management device driver 90 in the host computer 10 connected with the SCSI 25. The DVD-RAM drive 30, the PCMCIA connector 255, and the flash memory 250 are placed under the control of the storage device management device driver 90.

FIG. 16 shows the internal configuration of the storage device management device driver 90. The storage device management device driver 90 includes a file system interface software module 910 for realizing an interface with the file system software (11 in FIG. 17) of the host computer 10, and two surfaces provided in main storage in the host computer 10. A buffer control software module 920 for managing the buffers 951 and 952;
A DVD-RAM drive control software module 960 for controlling the M drive 30, a SCSI interface control software module 970 for controlling the SCSI interface, and a flash memory control software module 98 for controlling the flash memory 250
0.

The normal data flow will be described with reference to FIG. When the file system software 11 of the host computer 10 requests writing of one block of data, the file system interface software module 910 receives the data (17).
002), and is held in segment 17904 in buffer 951 (17005). The buffer control software module 920 is connected to the DVD-RA drive control software module 960 and the SCSI interface control software module 970.
An ECC block including the block is read from the M drive 30 to the buffer 952 (17006). Next, a segment 1790 is added to the read data of the ECC block.
4 are combined (17007). Then, ECC
DVD-RAM drive 3 for block data and ECC
Write to 0 (17010).

Referring to FIG. 18, a case where a failure occurs during writing to the DVD-RAM drive 30 will be described.
Writing to the DVD-RAM drive 30 (170
10) If a failure occurs at the time and the data cannot be written, the buffer control software module 9
20 writes the data of the ECC block in the buffer 952 to the flash memory 250 via the flash memory control software module 980 (1803)
0). Thereafter, even if a data write request is issued from the file system software 11 of the host computer 10, the buffer control software module 920
Refuse to receive data. When a request is made by the file system software 11 of the host computer 10 to read data of a block in the ECC block, if the data exists in the buffer 952, the data is transferred and the data is transferred to the buffer 952. If not, the data held in the flash memory 250 is transferred, and the data is not read from the DVD-RAM drive 30.

Also in the fourth embodiment, similarly to the first embodiment, a cache algorithm may be used, an SRAM with a battery may be used, or a non-stop power supply may be used.

According to the fourth embodiment, the storage device management device driver 90 is installed in the host computer 10.
Is provided, the effect of the first embodiment can be obtained with a simple configuration.

[0045]

According to the storage device management method and storage device management system of the present invention, write data smaller than the redundant data generation unit is received from the information processing device, and data lacking in the redundant data generation unit is read from the storage device. When generating redundant data after aligning the data in the redundant data generation unit and writing the data and the redundant data to the storage device, even if a failure occurs, all data aligned in the redundant data generation unit is not lost. Can be guaranteed.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a connection of a storage device management system according to a first embodiment.

FIG. 2 is a diagram illustrating an internal configuration of a storage device management system according to the first embodiment.

FIG. 3 is a diagram illustrating a flow of data in a normal state according to the first embodiment.

FIG. 4 is a diagram illustrating a data flow when writing fails due to a writing failure in the DVD-RAM drive according to the first embodiment.

FIG. 5 is a diagram illustrating a data flow when writing fails due to a reading failure in the DVD-RAM drive according to the first embodiment.

FIG. 6 illustrates a connection of a storage device management system according to a second embodiment.

FIG. 7 is a diagram illustrating an internal configuration of a storage device management system according to a second embodiment.

FIG. 8 is a diagram illustrating a connection of a storage device management system according to a third embodiment.

FIG. 9 is a diagram illustrating an internal configuration of a changer device according to a third embodiment.

FIG. 10 is a diagram illustrating an internal configuration of a storage device management system according to a third embodiment.

FIG. 11 is a diagram showing a flow of data in a normal state according to the third embodiment.

FIG. 12 shows one DVD-RA according to the third embodiment.
FIG. 14 is a diagram illustrating the first half of the data flow when writing fails due to a writing failure in the M drive.

FIG. 13 shows one DVD-RA according to the third embodiment.
FIG. 11 is a diagram illustrating the latter half of the data flow when writing fails due to a writing failure in the M drive.

FIG. 14 shows all DVD-RAs in the third embodiment.
FIG. 6 is a diagram showing a data flow when writing fails due to a writing failure in the M drive.

FIG. 15 illustrates a connection of a storage device management system (device driver) according to a fourth embodiment.

FIG. 16 is a diagram illustrating an internal configuration of a storage device management system (device driver) according to a fourth embodiment.

FIG. 17 is a diagram showing a flow of data in a normal state in the fourth embodiment.

FIG. 18 is a diagram illustrating a data flow when writing fails due to a writing failure in a DVD-RAM drive in the fourth embodiment.

FIG. 19 is a diagram showing a data management unit (block) and an ECC generation unit (ECC block) in a DVD-RAM drive.

FIG. 20 is a diagram showing a data flow in a conventional DVD-RAM device.

FIG. 21 is a diagram illustrating an internal configuration of a first modification of the storage device management system according to the first embodiment.

FIG. 22 is a diagram illustrating a configuration of a second modification of the storage device management system according to the first embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 ... Host computer, 11 ... File system software, 15 ... Fiber channel, 20 ... Storage device management system, 21 ... Storage device management system, 22 ... Storage device management system, 25 ... SCSI, 30 ... DVD-
RAM drive, 31 DVD-RAM device, 40 changer device, 80 changer device, 90 storage device management device driver, 210 fiber channel interface, 220 SCSI interface, 240
... Robot control interface, 250 ... Flash memory, 251 ... SRAM with battery, 255 ... PCMCIA
Connector, 259 PCMCIA controller, 280
... CPU, 282 ... Secondary cache, 284 ... Main memory,
286: Chipset, 299: Non-stop power supply, 40
5: changer SCSI interface, 410: robot control circuit, 420: storage, 450: media transport unit, 910: file system interface controller, 920: buffer control software module, 951: buffer, 952: buffer, 960 ...
DVD-RAM drive control software module,
970: SCSI interface control software module, 980: Flash memory control software module.

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5B018 GA04 HA03 KA14 LA01 MA16 NA06 QA05 QA15 5B065 BA04 CA03 CA11 CA30 CE01 EA03 EA12 EA23 EA24 EA26 EA39 ZA14 5B083 AA09 BB01 BB03 CC02 CD13 DD04 DD07 EE08

Claims (8)

[Claims] An information processing apparatus receives write data smaller than a redundant data generation unit, reads data lacking in the redundant data generation unit from a storage device, aligns the data in the redundant data generation unit, and generates redundant data. When the data and the redundant data are written to a storage device, and when the writing fails, the data aligned with the redundant data generation unit is held in a non-volatile storage unit, and the redundant data generation unit is A data read request for reading data belonging to the non-volatile storage means, wherein the data held in the non-volatile storage means is preferentially used and is not read from the storage device. 2. The method according to claim 1, wherein the redundant data is an ECC (Error Checker).
k Code) or RAID (Redundant Arrays of Inexpe)
2. The storage device management method according to claim 1, wherein the method is at least one of parity in nsive disks. 3. A write data receiving unit for receiving write data smaller than a redundant data generation unit from an information processing device, a data read unit for reading data insufficient for the redundant data generation unit from the storage device, Data writing means for generating redundant data after writing the data and writing the data and redundant data to the storage device, and a non-volatile storage for holding data aligned in the redundant data generation unit when the writing fails Means, and data priority use means for preferentially using data held in the nonvolatile storage means when there is a request for reading data belonging to the data of the redundant data generation unit from the information processing apparatus. A storage device management system, characterized in that: 4. The storage device management system according to claim 3, wherein the non-volatile storage means is a non-volatile memory or a battery-equipped memory that is detachable from the storage device management system. 5. A storage device management system comprising at least a time required to save data aligned in the write data management unit to the nonvolatile storage device when a normal power supply to the storage device management system is stopped. 5. The storage device management system according to claim 3, further comprising an emergency power supply device for backing up a power supply to the storage device. 6. RAID in combination with a plurality of storage devices
6. The storage device management system according to claim 3, wherein the storage device management system constitutes (Redundant Arrays of Inexpensive Disks). 7. The storage device management system according to claim 3, wherein the storage device management system is interposed between the information processing device and the storage device. 8. The storage device management system according to claim 3, wherein the storage device management system is installed inside one of an information processing device and a storage device.