JP2007257143A - Information storage system and information storage method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an inexpensive and highly reliable information storage system for storing information into a plurality of storage devices. <P>SOLUTION: This information storage system is provided with: a plurality of first storage means; a plurality of reading/writing means for reading/writing information for the first storage means; and a second storage means, wherein information to be stored in the first storage means is divided into a plurality of data blocks, and the data blocks are stored in the second storage means as a plurality of data columns distributed in parallel, and each data column is stored in each of the first storage means. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a system and method for dividing information into small units and distributing and storing the information in a plurality of storage devices, and further relates to a system and method for redundantly storing information in a distributed manner.

  In order to store a large amount of information, an information storage system containing a large number of information storage devices such as an optical disk and a hard disk is used. For example, Patent Document 1 discloses a storage system in which an autochanger apparatus that accommodates a plurality of media is further multiplexed. In such an autochanger device, a method of reading and writing in parallel to a plurality of information storage devices by a plurality of reading and writing means is adopted in order to improve the reading and writing speed. Further, a method (striping) is adopted in which information to be read / written is divided into smaller units and distributed and stored in a plurality of storage devices.

  Furthermore, in order to improve the reliability of the information storage system, a method of storing information in a redundant manner (striping with parity, RAID 5) is employed in addition to the above-mentioned striping. In this method, information divided into small units is made redundant and distributed and stored in a plurality of storage devices, so even if a failure occurs in one of the storage devices or a read error occurs, The lost information can be restored based on the information stored in the device. Patent Document 2 describes details of a disk array device that performs such redundancy.

JP-A-11-85409 JP-A-9-265714

  In general, in an information storage system that stores information in a distributed manner in a plurality of storage devices, it is necessary to provide the same number of read / write means as the number of storage devices that read and write in parallel at a time. For example, in the storage system described in Patent Document 1, N pairs of drives are mounted on N autochanger devices (FIG. 1), and in the disk array device described in Patent Document 2, five optical disks 221 to 221 are installed. Five optical disk drives 211 to 215 are mounted on 225 (FIG. 1).

  In such a conventional information storage system, if an attempt is made to improve the read / write speed, parallel read / write is performed using a larger number of read / write means, but this directly increases the size and cost of the apparatus. Will be connected. In addition, if any of a large number of reading / writing means breaks down, the reading / writing speed may be remarkably reduced or reading / writing may not be performed normally.

  The present invention has been made in view of such circumstances, and intends to provide a low-cost and highly reliable information storage system and information storage method of a method of storing information in a distributed manner in a plurality of storage devices. Is.

  As a result of diligent research in view of the above problems, the present inventor has provided auxiliary storage means as a pre-process that includes a relatively large capacity and high-speed auxiliary storage means such as a hard disk in an information storage system and is distributed to a plurality of storage devices. The information written in the means is divided into a predetermined number of data groups and temporarily stored, and these data groups are transferred to a predetermined number of read / write means and written to a predetermined number of storage devices, and further stored. We propose to solve the above problem by making the information stored in the apparatus redundant and ensuring the reliability by replacing the storage means when a reproduction error occurs.

  That is, the present invention provides an information storage system comprising a plurality of first storage means, a plurality of read / write means for reading and writing information from and to the first storage means, and a second storage means. Data to be stored in the second storage means as a plurality of data strings obtained by dividing the information to be stored into a plurality of data blocks and distributing the data blocks in parallel. An information storage system that generates a column and stores it in the second storage unit, and stores each data column and the redundant data column in each of the first storage units, wherein the read / write unit includes the first storage unit An information storage system is characterized in that an error is detected when information is read from the storage means, and the first storage means including the error is replaced with another first storage means for reading. It is intended to provide a Temu.

  The present invention also provides an information storage system comprising a plurality of first storage means, a plurality of read / write means for reading / writing information from / to the first storage means, and a second storage means, wherein the first storage means Obtained by dividing the information to be stored into a plurality of data blocks, generating a redundant data block for each predetermined number of the data blocks, and distributing the divided data blocks and the redundant data blocks in parallel A plurality of data strings stored in the second storage means, and each data string stored in each of the first storage means, wherein the read / write means stores information from the first storage means An information storage system, wherein an error is detected when reading, and the first storage means including the error is replaced with another first storage means for reading. It is intended to provide.

  In the information storage system of the present invention, the read / write means transfers the data block read from the first storage means to the second storage means, and the second storage means reconstructs information from the accumulated data blocks. It is characterized by doing.

  The present invention also performs information storage using an information storage device comprising a plurality of first storage means, a plurality of read / write means for reading / writing information from / to the first storage means, and a second storage means. A method, wherein information to be stored in the first storage means is divided into a plurality of data blocks, and the data blocks are stored in the second storage means as a plurality of data strings obtained by distributing in parallel. In addition, a data string in which the data string is made redundant is generated and stored in the second storage means, the data strings and the redundant data string are stored in the first storage means, and the read / write means Thus, an error is detected when information is read from the first storage means, and the first storage means including the error is replaced with another first storage means for reading. It is intended to provide a 憶方 method.

  The present invention also performs information storage using an information storage device comprising a plurality of first storage means, a plurality of read / write means for reading / writing information from / to the first storage means, and a second storage means. A method comprising: dividing information to be stored in the first storage means into a plurality of data blocks, generating a redundant data block for each predetermined number of the data blocks, and generating the divided data block and the redundancy A plurality of data strings obtained by distributing data blocks in parallel are stored in the second storage means, the data strings are stored in the first storage means, respectively, and the first read / write means An error is detected when information is read from the storage means, and the first storage means including the error is replaced with another first storage means for reading. It is to provide a method.

  In the information storage method of the present invention, the data block read from the first storage means is transferred to the second storage means by the read / write means, and the information is reconstructed from the accumulated data blocks by the second storage means. It is characterized by doing.

  According to the information storage system and the information storage method of the present invention, it is not necessary to provide the same number of read / write devices (read / write means) as the number of information storage devices (first storage units). Can be planned. In addition, information is redundantly stored, and an error correction unit in the case where an error exists in the stored information is provided, thereby providing a highly reliable information storage system.

  Further, in the information storage system adopting the conventional redundancy system such as RAID, the redundant information is always read at the time of reproduction. However, in the information storage system and the information storage method of the present invention, when a reproduction error occurs. Since only the storage means is replaced, the reproduction efficiency is improved while maintaining high reliability.

  The best mode for carrying out the information storage system and information storage method of the present invention will be described below in detail with reference to the accompanying drawings. 1 to 5 are diagrams illustrating embodiments of the present invention. In these drawings, the same reference numerals denote the same components, and the basic configuration and operation are the same. To do.

  FIG. 1 is a diagram schematically showing the system configuration of an information storage system according to an embodiment of the present invention. In FIG. 1, the information storage system includes a first storage unit 101, a second storage unit 106, and a control unit 105 that controls storage and reading of information in the first storage unit 101 and the second storage unit 106. Has been. The first storage unit 101 is a main storage unit for storing a large amount of information, and the second storage unit 106 is used as a temporary storage unit for assisting information storage in the first storage unit 101. It is what is done.

  The first storage unit 101 includes a plurality of recording media 102, a recording medium transport mechanism 103, and a plurality of read / write devices 104. In this information storage system, various media can be used as the recording medium 102. For example, it is possible to use optical media that perform recording and reproduction by heat and magnetomagnetism, magnetic media such as hard disks and floppy disks, semiconductor media such as ferroelectric memories and flash memories, and electronic memory media such as magnetoresistive memories. As the best mode of the present invention, it is assumed that write-once or rewritable DVD media (DVD-R, DVD-RAM, etc.) that are widely spread and have a large storage capacity and are inexpensive. The recording medium transport mechanism 103 transports the recording medium 102 and attaches / detaches it to / from the read / write device 104. The read / write device 104 is a device (drive) for performing recording / reproduction with respect to the recording medium 102. Note that the read / write device 104 need not be a device compatible with an information redundancy system such as RAID.

  The second storage unit 106 is composed of a rewritable storage element having a relatively high access speed and a large capacity, and is preferably composed of a hard disk device or the like.

  The control unit 105 includes a CPU, an MPU, and the like, and controls storage and reading of information in the first storage unit 101 and the second storage unit 106 according to a predetermined procedure according to a program (not shown).

  FIG. 2 is a diagram for explaining a striping method for information storage in the information storage system. In FIG. 2, information to be stored is first divided into data blocks of a predetermined length, reconstructed into four striped data strings, and redundant code data is generated from one data block of each string. . Each of the four striping data strings and one redundant code data string generated in this way is distributed and written on five recording media.

  In this information storage system, the operation procedure for performing such a write operation is as follows. First, the control unit 105 divides information input as a storage target in the information storage system into data blocks having a predetermined length (for example, 64 kilobytes), and stores the data blocks in the second storage unit 106 as four striping data strings. In addition, the control unit 105 generates a redundant code data sequence from these striping data and stores it in the second storage unit 106. Subsequently, the control unit 105 attaches the storage medium 102 to the read / write device 104 by the recording medium transport mechanism 103, transfers each data string stored in the second storage unit 106 to the predetermined read / write device 104, and transfers the data to the predetermined read / write device 104. It is written in the storage medium 102. This writing may be performed simultaneously by a plurality of read / write devices 104 or may be performed sequentially by one read / write device.

  In general, the writing operation to the storage medium 102 requires a certain amount of time. However, in this information storage system, the writing operation is completed when all necessary data is stored in the second storage unit 106. If the subsequent writing operation to the storage medium 102 is performed in the background, the user does not need to worry about the waiting time, leading to an improvement in work efficiency.

  FIG. 3 is a diagram for explaining a method of reading information in the information storage system. In FIG. 3, first, recording media (maximum of four) storing information to be read are loaded into the read / write devices 0 to 3, and each read / write device reads one data block in parallel. Here, an example of reading the written contents shown in FIG. 2 is shown, and “0000, 0001, 0002, 0003,”, “0004, 0005, 0006, 0007”, “0008” are sequentially read from the read / write devices 0 to 3. , 0009, 0010, 0011 "are read out, the original information is restored by connecting the data strings in one row.

  In this information storage system, the operation procedure for performing such a read operation is as follows. First, the control unit 105 identifies the recording medium 102 in which the information to be read is stored, and attaches them to the read / write device 104 by the recording medium transport mechanism 103. Thereafter, the data block group recorded on each recording medium 102 is read via each read / write device 104 and stored in the second storage unit 106. After reading all the data blocks, the data stored in the second storage unit 106 is integrated to restore the stored information.

  It should be noted that the control unit 105 can restore information from the data block group temporarily stored in the second storage unit 106 even if the reading operation in each reading / writing device 104 is not performed in synchronization. Is preferred. In order to realize this, for example, data blocks read from each read / write device 104 may be stored as separate data strings in the second storage unit 106.

  FIG. 4 is a diagram for explaining a method of performing reading while correcting errors in the recording medium in the information storage system. Consider a case where an error occurs in the shaded portion in the figure when information is read in the same manner as in the example shown in FIG. First, when it is detected that the data block “0004” is damaged, the recording medium 0 on which the data block is recorded is removed from the read / write device 0, and a recording in which a redundant code data string is recorded instead. The medium 4 is mounted on the read / write device 0. Since the data block “000B” can be read from the recording medium 4, the data block “0005”, “0006”, “0007” read from the recording media 1 to 3 is collated with the data block “000B”. The data block “0004” can be restored. Similarly, the damage to the data block “0010” can be restored by replacing the recording medium 2 with the recording medium 0. Note that after restoring damaged data, the mounting state of the recording medium may be restored or may not be restored.

  In this information storage system, for example, even if the read / write device is out of order or used for other purposes, only one read / write device can be used that is less than the number of recording media to be read simultaneously. Information can be read while replacing the recording medium every time the block is read. Alternatively, if all the data blocks to be read on the recording medium are read and stored in the second storage unit 106, then the recording medium is replaced and the reading is performed in the same manner, the time required for the recording medium replacement is reduced. Therefore, information can be read at high speed.

  The information storage system described above applies a striping method called “RAID 3” or “RAID 4” in the field of disk array technology. In addition, as shown in FIG. A striping method called “RAID5” that is distributed and stored in a medium can also be used. Alternatively, a striping method called “RAID0” in which only information is distributed and stored using a plurality of recording media without performing redundancy of information may be employed.

  While the information storage system and the information storage method of the present invention have been described with reference to specific embodiments, the present invention is not limited to these. A person skilled in the art can make various changes and improvements to the configurations and functions of the invention according to the above-described embodiments or other embodiments without departing from the gist of the present invention.

  The information storage system of the present invention can be configured as an information storage device including a plurality of optical disks, magnetic disks, magneto-optical disks, semiconductor storage elements, and the like. The information storage method of the present invention can be used as a method for reading and writing information in parallel in the information storage device.

1 is a diagram schematically showing a configuration of an information storage system of the present invention. It is a figure explaining the striping system for the information storage in the information storage system of this invention. It is a figure explaining the system which reads information in the information storage system of the present invention. It is a figure explaining the system which reads, correcting the error of a recording medium in the information storage system of this invention. It is a figure explaining the system which performs the information storage by RAID5 in the information storage system of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 ... 1st memory | storage part 102 ... Recording medium 103 ... Recording medium conveyance mechanism 104 ... Read / write device 105 ... Control part 106 ... 2nd memory | storage part

Claims (6)

A plurality of first storage means;
A plurality of read / write means for reading / writing information from / to the first storage means;
In an information storage system comprising second storage means,
The information to be stored in the first storage means is divided into a plurality of data blocks and stored in the second storage means as a plurality of data strings obtained by distributing the data blocks in parallel. Is generated and stored in the second storage means,
An information storage system for storing each data string and the redundant data string in each of the first storage means,
The read / write unit detects an error when reading information from the first storage unit, and reads the first storage unit including the error by replacing the first storage unit with another first storage unit. Storage system. A plurality of first storage means;
A plurality of read / write means for reading / writing information from / to the first storage means;
In an information storage system comprising second storage means,
The information to be stored in the first storage means is divided into a plurality of data blocks, a redundant data block is generated for each predetermined number of the data blocks, and the divided data block and the redundant data block are parallelized Are stored in the second storage means as a plurality of data strings obtained by being distributed to
An information storage system for storing each data string in each of the first storage means,
The read / write unit detects an error when reading information from the first storage unit, and reads the first storage unit including the error by replacing the first storage unit with another first storage unit. Storage system. The information storage system according to claim 1 or 2,
The read / write means transfers the data block read from the first storage means to the second storage means,
The information storage system, wherein the second storage means reconstructs information from accumulated data blocks. A plurality of first storage means;
A plurality of read / write means for reading / writing information from / to the first storage means;
A method for storing information using an information storage device comprising a second storage means,
The information to be stored in the first storage means is divided into a plurality of data blocks and stored in the second storage means as a plurality of data strings obtained by distributing the data blocks in parallel. Is generated and stored in the second storage means,
Storing each data string and the redundant data string in each of the first storage means;
An error is detected by the read / write means when reading information from the first storage means, and the first storage means including the error is replaced with another first storage means for reading. Memory method. A plurality of first storage means;
A plurality of read / write means for reading / writing information from / to the first storage means;
A method for storing information using an information storage device comprising a second storage means,
The information to be stored in the first storage means is divided into a plurality of data blocks, a redundant data block is generated for each predetermined number of the data blocks, and the divided data block and the redundant data block are parallelized Are stored in the second storage means as a plurality of data strings obtained by being distributed to
Storing each data string in each of the first storage means;
An error is detected by the read / write means when reading information from the first storage means, and the first storage means including the error is replaced with another first storage means for reading. Memory method. The read / write means transfers the data block read from the first storage means to the second storage means,
6. The information storage method according to claim 4, wherein information is reconstructed from the accumulated data blocks by the second storage means.

Images