JPH08263218A - Device and method for storing data - Google Patents

Abstract

PURPOSE: To reduce the number of times of access by reading compressed data corresponding to a designated physical block number while extending those data when storage in a memory is judged but reading data for the block unit of a logical block number when storage in a physical block is judged. CONSTITUTION: It is discriminated while using a logical block managing table whether the block data of the logical block number designated by a host processor are stored in a compressed data storage part or in a physical block storage part. When those data are recorded in the compressed data storage part, the read data are extended but when those data are stored in the physical block storage part, those data are read out as they are. Thus, the number of times of access to the physical block storage part is decreased and the access to the data storage device can be accelerated. Further, since the compressed data storage part stores data in a small compression size, storage capacity can be reduced.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data storage device and a data storage method for storing and reading data in a predetermined block unit on an exchangeable recording medium such as a plurality of optical disks.

[0002]

2. Description of the Related Art In recent years, the amount of data managed in an office has increased, and many large-capacity data storage devices have been proposed to manage such a large amount of data. For example, Japanese Examined Patent Publication (Kokoku) No. 4-38004 discloses a plurality of exchangeable recording media, their storages, and a drive unit for recording and reading data on and from the exchangeable recording media in order to realize a large capacity. , A data access method for a data storage device using an autochanger including an accessor unit for exchanging a removable recording medium between a storage unit and a drive unit is proposed.

An interface for accessing data on a storage device from a host processing device, which is an external device using the storage device, often uses a block (512 bytes or an integral multiple thereof) as an access unit, and the host processing device is a logical device. Data recording and reading are requested to the storage device using the specific block number as a parameter.

In Japanese Patent Publication No. 4-38004,
From the logical block number specified by the host, the number of the removable recording medium in which the data of the logical block is stored,
The conversion method to the physical block number on the exchangeable recording medium and the automatic exchange method of the exchangeable recording medium in which the data of the logical block is stored are described. According to this, the host processing device side can specify the data without being aware of which exchangeable recording medium the data is recorded on, and the processing at the time of data access is reduced.

In the example of Japanese Examined Patent Publication No. 4-38004, the physical block number is represented by the exchangeable recording medium number and the track number in the exchangeable recording medium, and the conversion from the logical block number to the physical block number is performed. , Is fixedly performed as in the following equation.

(Logical block number) / (number of tracks per medium) (1) exchangeable recording medium number = quotient of equation (1) track number = remainder of equation (1) Next, for example, a plurality of A conventional data storage method in a data storage device that stores data in predetermined blocks in an exchangeable recording medium will be described with reference to FIG.
It should be noted that the removable recording medium number and the track number are omitted here, and the physical block number is collectively managed in the physical block storage unit.

In FIG. 18, when the data storage device receives data in a predetermined block unit (block data) transmitted from the host processing device, it is designated by the host processing device and a host buffer for temporarily storing the block data. A logical block management table for converting a logical block number into a physical block number, and a physical block storage unit composed of a plurality of exchangeable recording media for storing block data. Each physical block in the physical block storage unit is designated by a physical block number. Also, assume that the block size is 1 Kbyte (K is 1024), the number of logical blocks managed by the data storage device is 10 M (M is 1024 K), and the total capacity is 10 Gbytes (G is 1024 M).

When reading the block data of the specified logical block number from the host processing device, the data storage device first obtains the physical block number corresponding to the specified logical block number from the logical block management table,
The block data of the block number is read from the physical block storage unit and transferred to the host buffer.

When writing block data to a logical block number designated by the host processing device, the data storage device first obtains a physical block number corresponding to the designated logical block number from the logical block management table, and then the host buffer The block data in the internal block is transferred to the physical block storage unit and stored therein in association with the physical block number.

[0010]

When a physical block storage unit is constructed using an exchangeable recording medium and an autochanger, a process of exchanging the exchangeable recording medium occurs when accessing a logical block, which results in extremely high performance. Becomes worse. A typical autochanger takes about 10 seconds to replace, which is about 3 digits slower than the access time of a hard disk drive or an optical disk drive. Therefore, a conventional cache has been used to reduce the number of accesses to the physical block storage unit, but it has not been sufficient to improve the performance.

Further, conventionally, the same number of physical blocks as the number of logical blocks is required before using the apparatus, and particularly when an exchangeable storage medium is used as a storage device of the physical block storage unit, it is equal to the number of logical blocks of the apparatus. It was necessary to prepare a new removable recording medium including the number of physical blocks from the beginning.

Therefore, according to the present invention, it is possible to reduce the number of times of access to the physical block storage unit having a low access speed, to read data at a higher speed, and to reduce the storage capacity of the physical block storage unit. An object of the present invention is to provide a data storage device and a data storage method.

[0013]

A data storage device according to the present invention stores a data by accessing a physical block for storing data of a block unit which is assigned in advance.
A data storage device for reading, wherein when storing the data in block units, a compression unit for compressing the data in block units, and a data length of the compressed data compressed by the compression unit is a predetermined value. When
First control means for controlling the compressed data to be stored in a memory that can be accessed at high speed; and the block when the data length of the compressed data compressed by the compression means is larger than a predetermined value. Second control means for controlling to store the unit data in the physical block, and when the block unit data is read, the designated block unit data is stored in either the memory or the physical block. Determining means for determining whether the data is stored in the memory, and when the determination means determines that the data is stored in the memory, the compressed data in block units stored in the memory is decompressed and read. When it is determined by the control means of No. 3 and the determination means that the physical block is stored in the physical block, And it includes a fourth control means for controlling so read the position data.

Further, the data storage device of the present invention manages data of a predetermined block unit by consecutive logical block numbers, and accesses a physical block storing the data of the block unit assigned to the logical block number. A data storage device for storing and reading the data, and compressing the block unit data to obtain compressed data when storing the block unit data of a specified logical block number; When the data length of the compressed data obtained by the compression means is equal to or less than a predetermined value, the compressed data is controlled to be stored in a memory that can be accessed at high speed in association with the designated logical block number. And the data length of the compressed data obtained by the compression means is a predetermined value. When it is also larger, the allocation unit that allocates the physical block to the specified logical block number, and the block unit data is stored in the physical block allocated to the specified logical block number by the allocation unit. And a second control means for performing control so as to read data in block units having a designated logical block number, based on the designated logical block number, the data in the block units is stored in the memory or the physical unit. Determining means for determining which of the blocks is stored, and when the determining means determines that the memory is stored in the memory, the compressed data corresponding to the specified physical block number is expanded from the memory. And a third control means for performing control so that the physical block is stored in the physical block. When it is determined, and a fourth control means for controlling so that reading data of the block unit from the physical block assigned to the designated logical block numbers.

Further, the data storage method of the present invention is a physical block in which data of a predetermined block unit is managed by a continuous logical block number, and the data of the block unit pre-assigned to the logical block number is stored. Is a data storage method for accessing and storing the data to read and write the data in the block unit of a designated logical block number, compressing the data in the block unit to obtain compressed data, When the data length of the compressed data is less than or equal to a predetermined value, the compressed data is stored in a memory that can be accessed at high speed in association with the designated logical block number, and the data length of the compressed data is When it is larger than a predetermined value, the data of the block unit is changed to the specified logical block number. When the data is stored in the physical block assigned to the block and the data in block units of the designated logical block number is read out, which of the memory and the physical block is stored based on the designated logical block number? When it is determined that the data is stored in the memory, the compressed data corresponding to the specified physical block number is decompressed and read from the memory, and it is determined that the data is stored in the physical block. Then, the data in the block unit is read from the physical block assigned to the designated logical block number.

Further, in the data storage method of the present invention, data in a predetermined block unit is managed by a continuous logical block number, and a physical block for storing the block unit data assigned to the logical block number is accessed. Is a data storage method of storing and reading the data, and storing the data in the block unit of a designated logical block number, compresses the data in the block unit to obtain compressed data, and the compressed data When the data length of the compressed data is less than or equal to a predetermined value, the compressed data is stored in a memory that can be accessed at high speed in association with the designated logical block number, and the data length of the compressed data is determined in advance. Is greater than the specified value, the physical block is assigned to the specified logical block number, When the data of the block unit is stored in the assigned physical block and the data of the block unit of the designated logical block number is read, the data of the block unit is stored in the memory based on the designated logical block number. Alternatively, it is determined in which of the physical blocks the data is stored, and when it is determined that the data is stored in the memory, the compressed data corresponding to the specified physical block number is decompressed and read from the memory,
When it is determined that the data is stored in the physical block, the data in the block unit is read from the physical block assigned to the designated logical block number.

Further, in the data storage method of the present invention, data in a predetermined block unit is managed by consecutive logical block numbers, and physical blocks pre-allocated to the logical block numbers on a plurality of exchangeable recording media. Is a data storage method for accessing and storing the data to read and write the data in the block unit of a designated logical block number, compressing the data in the block unit to obtain compressed data, When the data length of the compressed data is less than or equal to a predetermined value, the compressed data is stored in a memory that can be accessed at high speed in association with the designated logical block number, and the data length of the compressed data is When it is larger than the predetermined value,
When the data of the block unit is stored in the physical block assigned to the designated logical block number and the data of the block unit of the designated logical block number is read, based on the designated logical block number, It is determined whether the data is stored in the memory or the physical block, and when it is determined that the data is stored in the memory, the compressed data corresponding to the specified physical block number is decompressed from the memory. When it is determined that the data is read and stored in the physical block, the data in the block unit is read from the physical block assigned to the designated logical block number.

Further, according to the data storage method of the present invention, data in a predetermined block unit is managed by a continuous logical block number, and the physical block is assigned to the logical block number on a plurality of exchangeable recording media. A data storage method of accessing and storing and reading the data, wherein when storing the data in the block unit of a designated logical block number, the data in the block unit is compressed to obtain compressed data, and When the data length of the compressed data is less than or equal to a predetermined value, the compressed data is stored in a memory that can be accessed at high speed in association with the specified logical block number, and the data length of the compressed data is set in advance. When it is larger than a specified value, the physical block is allocated to the specified logical block number, and the allocation is performed. When the data of the block unit is stored in the designated physical block and the data of the block unit of the designated logical block number is read out, the data of the block unit is stored in the memory or the memory based on the designated logical block number. It is determined in which of the physical blocks the data is stored, and when it is determined that the data is stored in the memory, the compressed data corresponding to the specified physical block number is decompressed and read from the memory, and When it is determined that the data is stored in the physical block, the data in the block unit is read from the physical block assigned to the designated logical block number.

[0019]

When storing the block unit data of the specified logical block number, the block unit data is compressed to obtain the compressed data, and when the data length of the compressed data is equal to or less than the predetermined value, The compressed data is stored in a memory that can be accessed at high speed in association with the specified logical block number, and when the data length of the compressed data is larger than a predetermined value, the data in block units is When the data is stored in the physical block assigned to the designated logical block number and the data in block units of the designated logical block number is read out, based on the designated logical block number, the memory or the physical block When it is determined that it is stored in the memory, it is determined whether the memory is stored in the memory. When the compressed data corresponding to the specified physical block number is decompressed and read and it is determined that the compressed data is stored in the physical block, the block unit is determined from the physical block assigned to the specified logical block number. By reading the data of 1, the number of times of access to the storage unit configured by the physical block whose access speed is low can be reduced, and the data can be read at higher speed.

Further, when storing the block unit data of the designated logical block number, the block unit data is compressed to obtain the compressed data, and the data length of the compressed data is larger than a predetermined value. At this time, first, by allocating the physical block to the specified logical block number and storing the data in block units in the allocated physical block, the physical block allocated to the logical block number is: It suffices if the storage capacity is actually required, and therefore the storage capacity of the storage unit composed of physical blocks can be reduced.

[0021]

Embodiments of the present invention will be described below with reference to the drawings. First, the principle of the present invention will be described. FIG. 1 shows a logical block number (address) from an external host processing device in a data storage device including a physical block storage unit configured by a storage device with low access speed.
To explain the principle of the first data storage method for reducing the number of times of access to the physical block storage unit and realizing high-speed access processing when data is stored and read by being specified. belongs to.

Here, the block means a data amount of one unit of a predetermined data length, and the data is stored in the physical block storage unit in this block unit. Also,
A block in the physical block storage unit, that is, a physical block is managed by a physical block number and corresponds to each number of consecutive logical blocks designated by the host processing device. That is, on the host processing device side, block data is managed as consecutive logical block numbers as logical blocks, and the logical block number designated by the host processing device is converted into a physical block number in the data storage device before it is handled. A physical block in the physical block storage unit is accessed.

In the first data storage method shown in FIG. 1, block-unit data (hereinafter referred to as block data) transmitted from the host processing device with a logical block number specified is stored in the data storage device. At this time, first, the block data is compressed, and it is determined whether or not the size of the compressed data is smaller than a predetermined size. When the size is small, the compressed data is stored in the compressed data storage unit that can be accessed at high speed, and when the size is large, the non-compressed data is stored in the physical block storage unit as in the conventional case.

The storage devices of the compressed data storage unit and the physical block storage unit are all logical blocks (10M).
Shall have the capacity to store. The logical block management table has a flag area indicating whether each logical block is stored in the compressed data storage unit or the physical block storage unit, and is set at the time of writing the logical block.

The physical block number on the logical block management table is the number of the physical block corresponding to the logical block and is referred to when the flag is uncompressed. The compressed data number is the number of compressed data corresponding to the logical block, and is referred to when the flag is compressed.

When reading the block data of the specified logical block number from the host processing device, it is first determined whether the block data of the specified logical block number is stored in the compressed data storage unit or the physical block storage unit. The determination is made using the block management table, and the read data is decompressed when it is recorded in the compressed data storage unit, and is read as it is when it is recorded in the physical block storage unit.

By doing so, the number of accesses to the physical block storage unit is reduced, and the access speed in the data storage device can be increased. Moreover, since the compressed data storage unit stores data having a small compressed size, the storage capacity can be reduced.

The data compression / expansion method adopted here is
Use a high-speed one. For example, by adopting a run length code in units of bytes or words, it is possible to reduce the size after compression when all the data in the block are the same. In the formatting process at the start of use of the data storage device that stores data in block units, there are many processes that clear all blocks by 0, and even if a simple byte unit run length code is used, It is possible to reduce the number of accesses and improve the performance.

FIG. 2 shows the Run Length.
th) is for explaining the code. The run length code is based on the number of consecutive same values. In the example of FIG. 2, 6 1-byte values “00” are consecutive, and then 1-byte value “33” is 2 consecutive. This data consists of consecutive 8 bytes. In the data compression, the same value is converted into a continuous number and a value for each continuous section, and a code "00" indicating the end of the data is added at the end.

In FIG. 2, 8-byte data is compressed into 5-byte data. The longer the same value continues, the higher the compression rate. Next, the principle of the second data storage method will be described with reference to FIG.

This second data storage method is the same as the above first method.
Like all data storage methods, all the logical blocks (10M) are previously stored in the storage device of the physical block storage unit.
It is possible to change the storage device capacity for the physical block storage unit as needed, instead of allocating the capacity for storing the data. Therefore, a physical block management table and a storage device allocation releasing unit are added.

The physical block management table manages which physical block in the physical storage unit the actual storage device is assigned to and further stores the data of the logical block.

The storage device allocation release unit allocates and releases storage devices to physical blocks in the physical block storage unit. In FIG. 3, in the physical block storage unit, for example, of the 10M physical blocks, storage devices are allocated to the first 5M, and the latter 5M.
It is not assigned to M. The flag on the physical block management table of the physical block to which the storage device is not allocated is set to "unallocated". Of the physical blocks to which the storage device is assigned, the flag of the physical block that stores the uncompressed data of the block data is set to "Use", and the flag of the physical block that is not used for storing block data is It is set as "unused".

When storing the block data transmitted from the host processing device by designating the logical block number in the data storage device, first, the block data is compressed, and the size of the compressed data is a predetermined data size. Determine if it is less than size. When the size is small, the compressed data is stored in the compressed data storage unit that can be accessed at high speed, and when the previous data is recorded in the physical block storage unit, the flag of the physical block management table is changed from "Use" to "Not used". Change to "Use". If the compressed size is large, the physical block management table is searched for a flag whose flag is "unused", and uncompressed data is recorded in the physical block (physical block storage unit). When there is no physical block whose flag is “unused”, the storage device allocation releasing unit performs a process of adding a storage device for the physical block storage unit. When the number of physical blocks whose flag is “unused” increases, the storage device is released and the flag is set to “unallocated”.

As described above, in the storage device allocation releasing unit, the storage devices such as the removable recording medium can be effectively operated by increasing or allocating the storage devices for the physical block storage unit as necessary. Is possible.

Next, the first embodiment will be described with reference to FIG. FIG. 4 schematically shows the configuration of a data storage device that employs the above-described first data storage method.

The data storage device shown in FIG.
The block data of 1 Kbyte (1024 bytes) unit stores 10M (1024 × 1024) block data, and the total capacity is 10 G (10 × 1024 × 1024 × 1).
024) bytes.

In FIG. 4, the data storage device 1 is an LA
It is connected to the host processing apparatus 10 via a communication line 11 such as N. The data storage device 1 includes a CPU 2 and an HDD
3, main memory 4, communication control unit 5, host buffer 6, compression / decompression unit 7, first storage unit 8, second storage unit 9
Are connected to the system bus 12 with each other.

The HDD (hard disk drive) 3 is
The main memory 4 is a RA
It is composed of M and the like, and is for storing a control program of the entire data storage device 1 and data necessary for control.

The CPU 2 controls the entire data storage device 1, and reads the control program stored in the HDD 3 onto the main memory 4 at the start of operation and operates according to the control program.

The communication control unit 5 controls communication with the host processing unit 10 connected via the communication line 11, and an access command (for example, read command, write command) sent from the host processing unit. , The logical block number, the block data, etc., and the processing result is transmitted to the host processing apparatus 10.

The host buffer 6 temporarily stores block data. That is, at the time of a read command, the block data of the specified logical block number is read from the data storage device 1 (first storage unit 8 or second storage unit 9) and stored, and this data is stored by the communication control unit 5. It is transmitted to the host processing device 10. Also, at the time of a write command, the communication control unit 5 sends the write block data received from the host processing device 10 to the host buffer 6.
The data is temporarily stored in the data storage device 1 and then written in the data storage device 1 (the first storage portion 8 or the second storage portion 9).

The compression / expansion unit 7 is for compressing and expanding the block data. As a compression method, for example, there is a case of using a run length code in units of bytes.

The first storage unit 8 is, for example, a battery backup type memory, and is adapted to store a logical block management table. A compressed data storage unit is included in this logical block management table.

The second storage unit 9 is a storage device for physical block storage, which is composed of a storage medium with low access speed. FIG. 5 shows a storage example of the logical block management table in the first storage unit 8.

In this first embodiment, since there are a total of 10M logical blocks, the physical blocks are also fixedly set in advance so that 10M block data can be stored.

In FIG. 5, in the logical block management table, a flag, a physical block number, and an area for storing compressed data are assigned to all consecutive logical block numbers.

The flag indicates which of the first storage unit 8 and the second storage unit 9 the block data is stored in, and the flag column of the designated logical block number is "uncompressed". In the case of, the block data is stored in the second storage unit 9 in a non-compressed state (non-compressed data), and in the case of “compression”, the block data of the designated logical block number is compressed. The state (compressed data) is stored in the compressed data storage area of the logical block management table.

Here, in order to simplify the explanation, the compressed data consists of 1 byte, and when this value is "n", it indicates that all the bytes of the block data are "n". For example, when the block data cleared by 0 is written from the host processing device, the flag becomes “compressed” and the compressed data becomes “0”.

When the flag is "non-compressed", the non-compressed data is stored in the physical block in the second storage section 9 indicated by the physical block number stored in the physical block number column.

In FIG. 5, the block data whose logical block number is “0” is set to “uncompressed”, and the second storage unit 9
It means that the physical block number of the inside is stored in the physical block of "0".

The block data whose logical block number is "1" is set to "compressed", and is therefore stored in a compressed data storage area in the logical block management table of the first storage section 8 in a compressed state. It will be. The value of the compressed data is "0".

The data storage device 1 having such a configuration
Then, in order to collect block data of logical block numbers that are frequently accessed in a specific area of the second storage unit 9 and to shorten the seek time of the head, the correspondence between logical block numbers and physical block numbers is changed according to the access frequency. It is also possible to optimize it.

FIG. 6 is a flow chart for explaining the processing operation of the entire data storage device 1 of FIG. First,
The process proceeds to step S1, the initialization process of the data processing device 1 is performed, and thereafter, the access command transmitted from the host processing device 10 is received by the communication control unit 5, and the process according to each command is performed (step S2). In the following description, the logical block number is represented as LBN.

If the access command is "read" (step S3), the block data designated by the LBN is read from the host processing device 10 and the host processing device 1 is read.
Perform "reading process" to send to 0 (step S
4) Return to the command receiving process.

When the access command is "write" (step S5), "write processing" is performed to write the block data received from the host processor 10 to the physical block corresponding to the designated LBN (step S5).
6) Return to the command receiving process.

When the access command is "operation management" (step S7), "operation management processing" is performed (step S8), and the processing returns to the command reception processing. The operation management process includes an inquiry about the state of the data storage device 1, a format, and the like, which will not be described in detail here.

When the access command is "end",
The processing operation of the entire data storage device 1 is ended. Next, referring to the flowchart shown in FIG. 7, step S4 in FIG.
The reading process in step S1 will be described.

First, in step S10, L in the logical block management table stored in the first memory 8 is stored.
Reference the flag corresponding to BN. Flag is "uncompressed"
In the case of, the physical block number corresponding to the LBN in the logical block management table is extracted (step S11),
The block data of that block number is read from the second storage unit 9 and transferred to the host buffer 6 (step S
12).

When the flag is "compression", the LB in the logical block management table stored in the first storage unit 8 is used.
Extract the compressed data corresponding to N (step S1
3) The compression / decompression unit 7 decompresses the compressed data and transfers it to the host buffer 6 (step S1).
4).

Then, the communication control section 5 transmits the logical block data in the host buffer 6 to the host processing apparatus 10 (step S15), and returns the processing to the main program (step S2 in FIG. 6) (step S16).

Next, the write processing in step S6 of FIG. 6 will be described with reference to the flowchart shown in FIG. First, in step S20, the block data transferred from the host processing device 10 and received by the communication control unit 5 is transferred to the host buffer 6, and the compression / expansion unit 7 performs data compression processing (step S21).

Next, the size of the compressed data after block data compression is determined (step S22). If the size is larger than 1 byte, the LBN in the logical block management table stored in the first storage unit 8 is set. The corresponding physical block number is obtained (step S23), the block data (uncompressed data) in the host buffer 6 is written to the physical block corresponding to that physical block number in the second storage unit 9 (step S24), and the logical block The flag in the management table is set to "non-compressed" (step S25), the communication control unit 5 transmits the write end status to the host processing device 10 (step S26), and the process is executed by the main program (step S2 in FIG. 6). ) (Step S2)
7).

On the other hand, in step S22, if the size of the compressed data after data compression is 1 byte, the LB in the logical block management table stored in the first storage unit 8 is used.
One byte of compressed data is stored in the compressed data storage area corresponding to N (step S28), the flag is set to "compressed" (step S29), and the write end status is transmitted to the host processing device ( Step S26),
The process is returned to the main program (step S2 in FIG. 6) (step S27).

Next, the data compression processing of step S21 of FIG. 8 will be described with reference to the flowchart shown in FIG. For the sake of simplicity, the simplest data compression processing will be described.

First, in step S30, the variable i is set to 1
Of the first byte data in host buffer 6 (represented by host buffer [0] in FIG. 9) and i
The th byte data is compared (step S31).

If they are equal, the variable i is incremented by 1 (step S32), and the comparison is performed until 1024 or more (step S33). If all the byte data are equal to the start byte data, the compressed data is set to the start byte data (step S34), the compression size is set to 1 byte, and the process is a subprogram (step S22 in FIG. 8).
(Step S35).

If different byte data exists even for 1 byte (step S31), the compression size is made larger than 1 and the process is returned to the subprogram (step S22 in FIG. 8) (step S36).

As described above, according to the first embodiment, when the block data transmitted from the host processing device 10 is stored in the data storage device 1, the block data is first compressed and then stored. The size of the resulting compressed data is a predetermined size (here, 1
Byte) or less, the compressed data is stored in the first storage unit 8, and when the size of the compressed data is large, the compressed data is stored in the second storage unit 9, so that the second storage unit with slow access can be stored. It is possible to reduce the number of times of access to the unit 9, and thus a faster read operation is possible.

Next, the second embodiment will be described. FIG. 10 schematically shows the configuration of a data storage device that adopts the second data storage method described above. The same parts as those in FIG. 4 are designated by the same reference numerals and the description thereof will be omitted.

In FIG. 10, the data storage device 1 is L
It is connected to the host processing device 10 via a communication line 11 such as an AN. The data storage device 1 includes a CPU 2 and an HDD
3, a main memory 4, a communication control unit 5, a host buffer 6, a compression / decompression unit 7, a first storage unit 20, an accessor control unit 21, and an optical disk control unit 22 are connected to the system bus 12 and are configured. .

The data storage device 1 is a removable storage medium for storing 10M block data (total capacity is 10 Gbytes), for example, a second storage unit for storing block data by using a plurality of optical disks 24 and an autochanger. It is equipped with.

The optical disc 24 is capable of storing 1M block data by itself and uses a rewritable device such as a magneto-optical disc. The autochanger has 10 storage cells 25 and 1 optical disk drive 2
3 and an accessor 26. The storage cell is for storing an optical disc. The storage cells 25 are numbered “0” to “10”, the storage cells numbered “0” are windows for loading / unloading the optical disk 24 to / from the outside, and the 10 storage cells numbered “1” to “10”. The optical disc 24 can be stored in 25.

In the following description, a storage area composed of a plurality of optical disks is conceptually called a second storage section. The HDD 3, main memory 4, CPU 2, compression / expansion unit 7, communication control unit 5, and host buffer 6 are the same as those in FIG.

The first storage unit 20 is, for example, a battery backup type memory, and includes a logical block management table,
The physical block management table and other management tables are stored. Similar to FIG. 4, the logical block management table includes a compressed data storage unit.

The optical disk drive 23 is used for the optical disk 2
The optical disc drive control unit 22 controls the optical disc drive 23.

The accessor 26 moves the optical disk 24 between the storage cell 25 and the optical disk drive 23, and is controlled by the accessor controller 21. Incidentally, in FIG. 10, only five optical disks corresponding to half the total number of logical blocks are loaded in the autochanger. For each of the five optical disks 24,
"0" to "4" are added as optical disk numbers.

FIG. 11 shows a storage example of the logical block management table in the first storage section 8. FIG.
1, in the logical block management table, areas for storing flags, physical block numbers, and compressed data are assigned to all consecutive logical block numbers.

The flag indicates whether the block data is stored in the first storage unit 8 or the second storage unit, and "compression" is stored in the flag column of the designated logical block number. If so, the compressed data is stored in the column of compressed data in this table.

The compressed data consists of 1 byte. When this value is "n", all bytes of the block data are "n".
Is shown. For example, from the host processor 10 to 0
When the block data cleared by is written,
"Compressed" is stored in the flag field, and compressed data is "0".
Becomes

When the flag is "non-compressed", non-compressed data is stored in the physical block in the second storage section shown in the physical block number column. For example, the method for obtaining the optical disk number and the block number in each optical disk from the physical block number is as follows.

(Physical block number) / 1M (storage capacity per optical disc) (2) Optical disc number = quotient of formula (2) Block number = remainder of formula (2) Further, block data is compressed. Stored in the logical block management table, "-1" is set in the corresponding physical block number column. This indicates that no physical block is assigned to this logical block number.

As described above, in the first embodiment, the physical blocks are assigned to all the logical blocks, but in the second embodiment, the physical blocks are allocated to the second storage section as needed. By allocating, the number of used optical disks is saved.

FIG. 12 shows an example of storage of the physical block management table in the first storage section 8. This table indicates on which optical disk each physical block is allocated, and if so, whether the non-compressed data of the logical block is recorded. In the second embodiment, since five optical disks are loaded in the autochanger as the second storage unit, half of 5M physical blocks (physical block number is 10M) out of the total 10M logical blocks. Blocks on the optical disk are allocated to 0 to 5M-1).

A physical block whose flag in the physical block management table is "used" indicates that the block on the optical disk is allocated and the block data of the logical block is recorded. For example, in FIG. 12, the 0th block of the “0” th optical disc is allocated to the physical block having the physical block number “0”, and the physical block of FIG.
It can be seen from the logical block management table of 1 that the block data of the logical block of which LBN is 0 is stored.

The physical block whose flag is "unused" indicates that the block on the optical disk is allocated, but the block data of the logical block is not stored yet. A physical block whose flag is "unallocated" indicates that the block on the optical disk is not allocated.

In the flag column of the physical blocks having physical block numbers 5M to 10M-1 in the physical block management table, "unallocated" which means that the block on the optical disk is not allocated is stored.

FIG. 13 shows an example of storage of the optical disc management table in the first storage section 8. This table manages the storage cell number in the autochanger corresponding to each optical disk 24. In the second embodiment, the storage cell numbers of the five optical disks 24 are stored. When the optical disk 24 is removed from the optical disk drive 23, the storage cell 2 indicated by this number
Stored in 5. The storage cell number indicated by "-1" indicates that the optical disc 24 has not been inserted into the autochanger.

FIG. 14 is a storage example of the optical disk drive management table in the first storage section 8 for managing the numbers of the optical disks loaded in the optical disk drive 23.

The processing operation of the entire data storage device 1 of FIG. 10 is the same as that of FIG. Next, with reference to the flow chart shown in FIG. 15, the read processing of step S4 of FIG. 6 in the second embodiment will be described.

First, in step S40, the flag corresponding to the LBN in the logical block management table stored in the first storage unit 20 is referenced. If the flag is "non-compressed", the physical block number corresponding to the LBN in the logical block management table is extracted, and the above-mentioned formula (2) is used.
Based on the physical block number, the optical disc number and the block number within the optical disc are obtained (step S4).
1).

Then, referring to the optical disk management table and the optical disk drive management table stored in the first storage section 20, when the optical disk of the corresponding number is not loaded in the optical disk drive 23, the accessor control section The accessor 26 is driven under the control of the optical disc 21 and the optical disk 2 of the corresponding number is read from the corresponding storage cell 25.
4 is taken out and the optical disk 24 is loaded into the optical disk drive 23 (step S42).

Then, under the control of the optical disk drive controller 22, the optical disk drive 23 is driven to read the block data indicated by the physical block number from the corresponding optical disk 24, and the block data is transferred to the host buffer 6 (step S43). .

If the flag is "compressed" in step S40, the compressed data corresponding to the LBN in the logical block management table is extracted (step S44), decompressed by the compression / decompression unit 7, and then stored in the host buffer 6. Transfer (step S45).

Then, the block data in the host buffer 6 is transmitted to the host processor 10 by the communication controller 5 (step S46), and the process is returned to the main program (step S2 of FIG. 6) (step S47).

Next, the write processing of step S6 of FIG. 6 in the second embodiment will be described with reference to the flow chart shown in FIG. First, the block data transmitted from the host processing device is received by the communication control unit 5, transferred to the host buffer 6 (step S50), and the compression / expansion unit 7 performs data compression processing (step S51). The data compression processing at this time is, for example, that shown in FIG. Then, the size of the compressed data after data compression is determined (step S52).

If the size is larger than 1 byte, the logical block management stored in the first storage unit 20 is checked in order to check whether the physical block is already allocated corresponding to the logical block number indicated by LBN. The physical block number corresponding to the LBN on the table is obtained (step S53).

At this time, if the value of the storage area of the physical block number is "-1" (step S54), the physical block is not allocated, so the physical block allocation processing is performed (step S55), and then the logical block is allocated. The physical block number assigned to the physical block number corresponding to the LBN on the management table is set (step S56),
"Use" is set to the flag corresponding to the physical block number on the physical block management table (step S57).

If the value of the storage area of the physical block number is not "-1" in step S54, the physical block has already been allocated, so the physical block allocation process of step S55 is skipped and the process proceeds to step S58.

Next, based on the equation (2), the optical disk number and the block number in this optical disk are obtained from the physical block number (step S58), and if this optical disk is not loaded, it is loaded (step S58). S59).
Then, the optical disk drive 23 is driven to write the uncompressed data in the host buffer 6 to the block indicated by the block number (step S60). After that, the flag corresponding to the LBN on the logical block management table is set to "non-compressed" (step S61), and the communication control unit 5 transmits the write end status to the host processing device 10 (step S62). It returns to the program (step S2 of FIG. 6) (step S63).

In step S52, if the size of the compressed data after data compression is 1 byte, the LB on the logical block management table is checked to see if a physical block has already been allocated to the block data indicated by LBN.
Obtain the physical block number corresponding to N (step S6)
4). If the value of the storage area of the physical block number is other than "-1" (step S65), since the physical block is allocated, the physical block number corresponding to the LBN on the logical block management table is released to release the physical block. Is set to "-1" (step S66), and "unused" is set to the flag corresponding to the physical block number on the physical block management table (step S67). afterwards,
One byte of compressed data is set in the compressed data storage area corresponding to the LBN in the logical block management table (step S68), and the flag is set to "compressed" (step S68).
69), the write end status is transmitted to the host processing device (step S62), and the process is returned to the main program (step S2 in FIG. 6) (step S63).

Next, the physical block allocation processing in step S55 of FIG. 16 will be described with reference to the flowchart shown in FIG. First, the physical block management table stored in the first storage unit 20 is searched for a physical block whose flag is "unused" (step S7).
0). If found (step S71), the assigned physical block number is used as the physical block number (step S76), and the subprogram (step S56 in FIG. 16) is executed.
(Step S76).

If the optical disc is not found in step S71, the operation manager is instructed to insert a new optical disk into the autochanger, and the operator is asked to insert the optical disk from the entrance / exit of the autochanger (step S72). And
The row having the storage cell number “−1” on the optical disc management table stored in the first storage unit 20 is selected, and the optical disc number is set in the optical disc number column corresponding to the storage cell number. The new optical disk at the entrance / exit is moved to the storage cell 25 indicated by the storage cell number (step S73). Since a storage device for physical blocks for one optical disk has been added, the capacity for one optical device from the top row where the flag on the physical block management table is "unallocated" is 1
Allocating physical blocks for M and setting the flag to "unused" (step S74), the flag is set to "unused".
The number of the physical block is returned to the subprogram (step S56 in FIG. 16) (steps S75 and S76).

As described above, according to the second embodiment, when the block data transmitted from the host processing device 10 is stored in the data storage device 1, the block data is first compressed and then stored. The size of the resulting compressed data is a predetermined size (here, 1
It is possible to save the optical disk to be used by allocating the physical block on the optical disk 20 as the second storage unit only when the physical block is larger than the (byte). In particular, when the operation of the data storage device 1 is started, the operation can be started even if there are not 10 optical disks to be loaded.

As described with reference to FIG. 9, in the data compression processing, if all the data in the block data have the same value, it is possible to reduce the size after compression to several bytes. Although it is used, not limited to this, even if a run length code of byte or word is adopted,
It is effective to adopt other compression methods. However, it is desirable to adopt the fastest possible data compression / decompression method. Particularly, in the data storage device that stores and reads data in block units as in the first and second embodiments described above, in the formatting process at the start of device operation, the process of clearing all blocks to 0 is required. Even with many simple byte-length run length codes, it is possible to reduce the number of accesses to the physical block storage unit and improve the performance.

In the block data writing process, when the compressed sizes are compared, the reference data size is defined as 1 byte in the first and second embodiments, but the present invention is not limited to this case. It can be appropriately defined according to the size of the data or the data compression processing.

[0107]

As described above, according to the present invention,
It is possible to provide a data storage device and a data storage method capable of reducing the number of times of access to a physical block storage unit having a low access speed, reading data at a higher speed, and reducing the storage capacity of the physical block storage unit.

[Brief description of drawings]

FIG. 1 is a diagram for explaining the principle of a first data storage method of the present invention.

FIG. 2 is a diagram for explaining a run length code.

FIG. 3 is a diagram for explaining the principle of the second data storage method of the present invention.

FIG. 4 is a diagram schematically showing a configuration of a data storage device according to a first embodiment of the present invention.

FIG. 5 is a diagram showing an example of storage of a logical block management table.

FIG. 6 is a flowchart for explaining the overall processing operation of the data storage device of FIG.

FIG. 7 is a flowchart illustrating a read process.

FIG. 8 is a flowchart for explaining a writing process.

FIG. 9 is a flowchart illustrating a data compression process.

FIG. 10 is a diagram schematically showing a configuration of a data storage device according to a second embodiment of the present invention.

FIG. 11 is a diagram showing a storage example of a logical block management table.

FIG. 12 is a diagram showing a storage example of a physical block management table.

FIG. 13 is a diagram showing a storage example of an optical disc management table.

FIG. 14 is a diagram showing a storage example of an optical disc drive management table.

FIG. 15 is a flowchart illustrating a read process.

FIG. 16 is a flowchart for explaining a writing process.

FIG. 17 is a flowchart illustrating a physical block allocation process.

FIG. 18 is a diagram for explaining the principle of a conventional data storage method.

[Explanation of symbols]

1 ... Data storage device, 2 ... CPU, 3 ... Hard disk drive (HDD), 4 ... Main memory, 5 ... Communication control unit, 6 ... Host buffer, 7 ... Compression / decompression unit, 8 ... First
Storage unit, 9 ... second storage unit, 10 ... host processing device,
11 ... communication line, 12 ... system bus.

Claims (6)

[Claims] 1. A data storage device that stores and reads the data by accessing a physical block that stores data of a block unit allocated in advance, wherein the block unit is stored when the data of the block unit is stored. And a compression means for compressing the data, and performing control to store the compressed data in a memory that can be accessed at high speed when the data length of the compressed data compressed by the compression means is equal to or less than a predetermined value. No. 1 control means, and second control means for performing control so as to store the data in block units in the physical block when the data length of the compressed data compressed by the compression means is larger than a predetermined value. When reading data in block units, the specified data in block units is stored in the memory or the physical block. And a determination means for determining which of the stored data is stored in the memory, and when the determination means determines that the data is stored in the memory, the compressed data in block units stored in the memory is expanded and read. A third control means for controlling to output the data, and a fourth control means for controlling to read the data in block units from the physical block when the determination means determines that the data is stored in the physical block. A data storage device comprising: a control unit. 2. Data of a predetermined block unit is managed by consecutive logical block numbers, and a physical block for storing the data of the block unit allocated to the logical block number is accessed to store and read the data. A data storage device for performing, when storing data of the block unit having a designated logical block number, compressing the data of the block unit to obtain compressed data, and a compression unit obtained by the compression unit. When the data length of the data is equal to or less than a predetermined value, the compressed data is associated with the designated logical block number and is controlled to be stored in a memory that can be accessed at high speed. When the data length of the compressed data obtained by the compression means is larger than a predetermined value, the specified Allocation means for allocating the physical block to the logical block number, and second control for storing the data in block units in the physical block allocated to the specified logical block number by the allocation means When the control unit and the data in the block unit of the designated logical block number is read, whether the data in the block unit is stored in the memory or the physical block based on the designated logical block number. And a control means for decompressing and reading compressed data corresponding to the specified physical block number from the memory when the determination means determines that the data is stored in the memory. When the third control means and the determination means determine that the data is stored in the physical block, the specified A fourth control means for performing control so as to read the data in block units from the physical block assigned to the assigned logical block number, the data storage device. 3. Data of a predetermined block unit is managed by consecutive logical block numbers, and a physical block for storing the data of the block unit pre-allocated to the logical block number is accessed to store the data. , A data storage method for reading, wherein when storing the data in block units of a specified logical block number, the data in block units is compressed to obtain compressed data, and the data length of the compressed data is set in advance. When the value is less than or equal to a predetermined value, the compressed data is stored in a memory that can be accessed at high speed in association with the specified logical block number, and the data length of the compressed data is larger than a predetermined value. At this time, the data of the block unit is transferred to the physical block assigned to the specified logical block number. When reading the data of the block unit of the designated logical block number, it is determined whether the memory is stored in the memory or the physical block based on the designated logical block number, and the data is stored in the memory. When it is determined that it is stored, the compressed data corresponding to the specified physical block number is decompressed and read from the memory, and when it is determined that it is stored in the physical block, the specified data is stored. A data storage method comprising reading the data in block units from a physical block assigned to the logical block number. 4. Data of a predetermined block unit is managed by consecutive logical block numbers, and a physical block for storing the data of the block unit allocated to the logical block number is accessed to store and read the data. A method of storing data, wherein when storing the data in block units of a designated logical block number, the data in block units is compressed to obtain compressed data, and the data length of the compressed data is predetermined. When the value is less than or equal to a value, the compressed data is stored in a memory that can be accessed at high speed in association with the designated logical block number, and when the data length of the compressed data is larger than a predetermined value, the The physical block is assigned to a specified logical block number, and the assigned physical block is assigned to the physical block. When storing the data of the lock unit and reading the data of the block of the designated logical block number, the data of the block unit is stored in either the memory or the physical block based on the designated logical block number. When it is determined that it is stored in the memory, the compressed data corresponding to the specified physical block number is decompressed and read from the memory and stored in the physical block. When it is determined that the data is stored, the data in block units is read from the physical block assigned to the designated logical block number. 5. Data of a predetermined block unit is managed by consecutive logical block numbers, and physical blocks previously assigned to the logical block numbers on a plurality of exchangeable recording media are accessed to store the data. , A data storage method for reading, wherein when storing the data in block units of a specified logical block number, the data in block units is compressed to obtain compressed data, and the data length of the compressed data is set in advance. When the value is less than or equal to a predetermined value, the compressed data is stored in a memory that can be accessed at high speed in association with the specified logical block number, and the data length of the compressed data is larger than a predetermined value. At this time, the data of the block unit is stored in the physical block assigned to the designated logical block number, and designated. When reading the data in block units of the specified logical block number, it is determined whether the data is stored in the memory or the physical block based on the specified logical block number, and the data is stored in the memory. When it is determined that the compressed data corresponding to the specified physical block number is decompressed and read from the memory, and when it is determined that the compressed data is stored in the physical block, the specified logical block number A data storage method comprising reading the data in block units from a physical block assigned to the block. 6. Data of a predetermined block unit is managed by consecutive logical block numbers, and physical blocks allocated to the logical block numbers on a plurality of exchangeable recording media are accessed to store the data, A data storage method for reading, wherein when storing data in the block unit of a specified logical block number, the data in the block unit is compressed to obtain compressed data, and the data length of the compressed data is predetermined. When the compressed data is equal to or smaller than a predetermined value, the compressed data is stored in a memory that can be accessed at high speed in association with the specified logical block number, and the data length of the compressed data is larger than a predetermined value. , Assigning the physical block to the specified logical block number, and assigning the block to the assigned physical block. When storing the data of a unit and reading the data of the block of the specified logical block number, the data of the block unit is stored in either the memory or the physical block based on the specified logical block number. If it is stored in the memory, the compressed data corresponding to the designated physical block number is decompressed and read from the memory and stored in the physical block. When it is determined that the data is stored in block units, the data is read from the physical block allocated to the designated logical block number.