KR100513814B1 - Data management device, data management method and storage medium - Google Patents

Abstract

The distributed management information, such as the distributed management information 0 for managing data for each predetermined unit, is stored in the redundant area of each block of the flash memory, while the collective management information for collectively managing the data is stored in a predetermined block of the flash memory. . When using flash memory, first, it is detected whether the set management information is an error. If an error is detected, the centralized management information is constructed from the distributed management information in the redundant area of each block. When the data is changed, the aggregate management information is reconstructed based on the distributed management information in the redundant area of each block, and the aggregate management information is stored in the predetermined block.

Description

Data management device, data management method and storage medium

The present invention relates to a data management apparatus using a flash memory, a data management method and a storage medium.

Conventionally, when an electronic device such as a computer is used as the main storage device, the memory device has been used with sufficient capacity of this main storage device. Examples of this storage device include magnetic tapes, magnetic disks, magneto-optical disks, paper tapes, and the like. Among these devices, memory cards have been widely used for transfer rates of relatively high speed.

In addition, this memory card is incorporated in cameras of still image imaging devices (so-called still cameras) and video tape recorders so as to be used as storage devices for recording imaging information.

The memory card includes embedding a storage means such as a semiconductor integrated circuit in a box made of synthetic resin, for example. As a semiconductor integrated circuit, a flash memory (electrically erasable program ROM (Read Only Memory) or the like is used.

By the way, in a storage device such as the memory card, data management is executed by a so-called distributed management method or collective management method.

The distributed management method is executed as follows. The storage means for storing data in the storage device is divided into data erasing units such as blocks, and provides block management information such as block flags, logical addresses, and connection information for each block, and is distributed and created for each block. . This management information list will hereinafter be referred to as distributed management information. In the distributed management method, in order to create a collective management table that can collectively execute management of the storage means of the storage device when the electronic device is activated to read and / or write data to the storage device, firstly. Then, the electronic device collects the distributed management information list distributed in each block of the storage means, reads it out into the storage means of the electronic device, and thus executes data management of the storage device.

In the case of using this distributed management method, the data in each block in the storage means of the storage device is changed by transmission in the electronic device, and the distribution management information is also changed. Therefore, each time the electronic device is activated as described above, the latest distributed management information list is collected in each block of the storage means in the storage device in order to create a management table in the storage means of the electronic device. That is, according to this distributed management method, in the example, even if some blocks in the storage means in the storage device are broken, at the next startup of the electronic device, the damaged blocks for creating the collective management table in the storage means of the electronic device. Collect a list of distributed management information that has been deleted. Therefore, in the distributed management method, even when some blocks in the storage means of the storage device are damaged, data can be written and / or data can be read for the unbroken block. That is, since there is no case where writing and / or reading is not performed for all the storage means, the distributed management method is relatively resistant to damage.

However, when the distributed management method is used for a storage device having a large capacity and a large number of blocks, and activates an electronic device that writes and / or reads data in response to the storage device, a large amount of time is required to create a collective management table in the electronic device. Takes Therefore, a large amount of time is required at the start of the electronic device, which hinders high speed processing.

Another method, the set management method, is executed as follows. A list of management information such as block flags, logical addresses, and connection information of each block in the storage means in the storage device is stored in one place in the storage means and created as management information for collectively managing the data of each block. Hereinafter, the management information is referred to as collective management information. When activating an electronic device for reading and / or writing data to the storage device, the electronic device reads out the collection management information to create a collection management table in the storage means of the electronic device, and thus the storage device. Perform data management.

In the case of using this collective management method, when writing data and / or reading data to and from the storage device, the electronic apparatus needs to read the collective management information, so that the electronic apparatus is performed in a short time, High speed processing can be performed.

However, in the case of managing the data of the storage device by the above-described collection management method, if the collection management information is broken in the storage means of the storage device, it is impossible to perform data writing and / or reading of all the storage means. This method is vulnerable to data corruption. In addition, according to the set management method, whenever the data of each block is changed in the storage means of the storage device, it is necessary to change the set management information as a whole for matching data between the set management information and each block. In order to rewrite a large number of blocks, a large amount of time is required to change the collective management information, thus preventing the high speed processing.

It is therefore an object of the present invention to provide a data management apparatus, a data management method, and a recording medium that are resistant to damage and enable high-speed processing of data.

The data management apparatus according to the present invention stores arbitrary data in a data storage area configured for each predetermined storage unit, and stores distributed management information for managing data for each predetermined storage unit in the distributed management information storage area, and sets management. If it is determined that the storage means for storing the collective management information for collectively managing data according to the distributed management information in the information storage area and the management means for determining at the time of startup whether the collective management information is valid and the collective management information are effective, the collective management information And data management in accordance with the distributed management information when the collective management information is invalid.

The data management method according to the present invention comprises the steps of: storing arbitrary data in a data storage area configured for each predetermined storage unit and storing and managing distributed management information for managing data for each predetermined storage unit in the distributed management information storage area. Storing the collective management information for collectively managing data in accordance with the distributed management information in the management information storage area; determining whether the collective management information is valid at startup; and if the collective management information is determined to be valid, the data based on the collective management information. Or managing data based on the distributed management information if it is determined that the set management information is invalid.

The storage medium according to the present invention is characterized by storing arbitrary data in a data storage area configured for each predetermined storage unit, and storing distributed management information for collectively managing data in accordance with the distributed management information in the collective management information storage area.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. The present invention is applied to the memory card system 1 including the host computer 10 and the memory card 20, as shown in the example of FIG.

Although description is made here in the case of writing the image data transmitted from the host computer 10 to the memory card, the same applies to the case of other data including audio data.

As shown in Fig. 1, the host computer 10 temporarily stores image data and the like on a hard disk drive 11 (hereinafter referred to as an HDD) that stores still image image data and audio data, and the HDD 11; And a display for displaying an image on the basis of image data supplied through a random access memory (RAM) 12 and a display interface 13 (hereinafter referred to as display I / F) and display I / F 13 that are read out. Serial interface 15 (hereinafter referred to as serial I / F) for performing data transmission and reception to and from memory card 20 through three lines 14 and CPU 17 for controlling the bus 16 and the entire system. (Central Processing Unit) and a ROM 18 including a program for controlling the CPU 17.

The RAM 12 temporarily stores audio data and image data stored in the HDD 11 via the bus 16, for example, and stores image data via the bus 16 as necessary. It supplies to F15.

The display 14 not only supplies the image data read out from the HDD 11 via the bus 16 and the display I / F in order to display still images based on these image data, but also the image data in the memory card 20. Is supplied.

The serial I / F 15 transmits image data to the memory card 20 via three lines and receives image data in the memory card. Specifically, the serial I / F 15 transmits image data and control data for writing to the memory card 20 and receives image data read out from the memory card 20 through the first line. The serial I / F 15 outputs a status signal indicating the switching state in accordance with the switching between the image data and the control data on the first line through the second line. In addition, the serial I / F 15 transmits a serial clock SCLK for transmitting the control data and the image data through the third line.

The CPU 17 not only reads the image data from the RAM 12 and the HDD 11 and writes the image data to the RAM 12 or the like, but also controls transmission and reception of the image data from the memory card 20. In the example, the CPU 17 issues data write to the memory card 20 by designation of an address.

As shown in Fig. 2, the memory card 20 of this embodiment uses a flash memory (A) as a storage means for storing control data and control IC 21 for receiving image data and received image data in the host computer 10. 22) (electrically erasable program ROM).

Specifically, the control IC 21 stores image data from a serial / parallel-parallel / serial interface sequencer (hereinafter referred to as a sequencer) 31 and a sequencer 31. A flash interface sequencer 33 (hereinafter referred to as a flash I / F sequencer) for supplying image data from the page buffer 32 and the page buffer 32 to the flash memory 22 for temporary storage and error correction processing An ECC encoder / decoder 34, a command generator 35 for generating a predetermined control command, a configuration ROM 36 containing conversion information, and an oscillator 37 for supplying a clock to each circuit. do.

The sequencer 31 is connected to the serial I / F 15 of the host computer 10 via the first to third lines. Accordingly, the sequencer 31 is supplied with serial data DIO composed of image data and control data from the host computer 10 as well as the status signal and the serial clock SCLK.

The sequencer 31 converts the serial data DIO supplied from the host computer 10 into parallel data in synchronization with the serial clock SCLK. Of the parallel data, the sequencer 31 supplies control data to the command generator 35 and image data to the page buffer 32 in the example.

The page buffer 32 is a buffer memory that stores image data supplied from the sequencer 31 for each page. Error correction is added to the picture data stored in the page buffer 32 by an ECC encoder / decoder. The page buffer 32 supplies the image data of one page having an error correction code to the flash memories 22a to 22d via the flash I / F sequencer 33. Therefore, image data is written from the host computer 10 into the flash memories 22a to 22d.

In addition, the image data read out from the flash memories 22a to 22d is supplied to the page buffer 32 through the flash I / F sequencer 33.

The page buffer 32 stores image data from the flash I / F sequencer 33. At this time, the ECC encoder / decoder 34 executes error correction processing based on an error correction code added to the data stored in the page buffer 32.

The page buffer 32 reads out error corrected data every page and supplies the data to the sequencer 31. The sequencer 31 converts the parallel image data supplied from the page buffer 32 into serial data DIO and transmits it to the host computer 10.

The command generator 35 generates a control command based on the control data in the sequencer 31. In addition, the command generator 35 writes a busy command (hereinafter referred to as a busy signal) indicating the state in which image data is written to the flash memory 22 or the image data is being read from the flash memory 22. A command is sent to the host computer 10 via the sequencer 31. The host computer 10 receives a busy signal and a ready signal to recognize an operating state of the memory card 20.

The configuration ROM 36 includes translation information of the memory card 20 and information of initial setting values. Therefore, when the host computer 10 and the memory card 20 are connected, the command generator 35 first reads the translation information and the like from the configuration ROM 36 through the sequencer 31 and based on this information. A predetermined command is issued to execute predetermined initial setting of the memory card 20.

Here, the flash memory 22 is NAND type. As shown in Fig. 3, the flash memory 22 is composed of a plurality of blocks including two boot areas, a data area made up of a plurality of blocks, and an area having a plurality of block set management information (set management files). .

The boot area includes one block each of a boot block and a boot block backup. The boot block includes the address of the block that initially contains the data read, for example, data for recognizing the block and aggregate management information. In addition, the bootblock backup is, in the example, a copy of the bootblock for use in case the bootblock crashes.

The data area is used to store various kinds of data such as image data and audio data. In the area including the collective management information, the collective management information that forms the data for managing the information necessary for access is stored.

In each block, one page (= 528 bytes) consisting of a main data area storing 512 bytes of main data and a redundant area of 16 bytes is composed of a predetermined number of pages. The redundant area stores distributed management information indicating the storage status of the page and the like.

Fig. 4 showing distributed management information of each page shows block permission / non-availability information (1 byte), block flag (1 byte), block last flag (4 bytes), reference flag (4 bits), management flag (1 byte), Logical address (2 bytes), connection address (2 bytes), format reservation (3 bytes), distributed information ECC (2 bytes), and data ECC (3 bytes).

"Block enable / disallow information" is information for identifying whether a block is in an unusable state. In an example, when a block is broken and unavailable, the "block grant / disallow information" is overwritten to indicate that the block is unavailable. The "block flag" indicates data not written in a block, that is, an unused block, a block used as a file head, or a block used outside the file head, or when this block becomes unnecessary due to erase or rewrite. Indicates that the block is set to the used state to be erased later. The "block final flag" indicates whether a block is the last block when a plurality of blocks are composed of one file. When the "block final flag" indicates that the block is the last block, the "connection address" is invalid even if specified. "Reference Flag" is a flag that specifies a reference of additional information described later. This additional information exists in the last page of the block. The "management flag" is 1 byte of information. Three bits (bits 2 through 0) of one byte are stored in the set management information table, while the remaining five bits are not stored in the set management table but are used for error correction processing and the like.

"Logical address" indicates the logical address of the block. "Connection address" indicates the logical address of the block connected to this block. When the block is already distinguished as the last block, the "block final flag" is turned on and "0xffff" is set in the "connection address". When it is impossible to tell whether this block is the last block, a logical address is secured and set to the "connection address". When this block becomes the last block, the overlap is executed and the “block final flag” is turned on. "Form reservation" is 3 bytes of information used as a reservation area. "Distributed information ECC" is used to correct errors in "management flag", "logical address", "connection address" and "format reservation". "Data ECC" is used to correct data errors other than distributed management information.

The distributed management information on the last page of each block is used as additional information. The additional information includes a redundant base area and an additional management information area, as shown in FIG. The redundant base area is the same as the above distributed management information. The additional management information area includes an identification number (1 byte), an effective data size (2 bytes), and a format reservation (5 bytes). The "identification number" is used to overwrite a file to identify the original file and the destination when updating using the same logical block.

When a new logical address is used, "0" is described, and when updated, its value is increased. "Effective data size" indicates the size of valid data in the block. In the example, when the block has no free area, it is described as "0xffff" and the "reference flag" is turned on and describes the effective data size-1 of the block.

In addition, the set management information is constructed based on the distributed management information of each block. As shown in Fig. 6, the set management information is composed of a header, a bitmap table, an address conversion table and a connection information table, which contain start position information for searching for a free area in each block of the flash memory 22. The header includes an OS header, filehead, system entry list, and system information.

As shown in Fig. 7, the OS header includes file ID (2 bytes), file translation (2 bytes), file size (4 bytes), number of blocks used (2 bytes), number of connections (1 byte), and date (8). Byte), author / model code (4 bytes), initial registration directory number (2 bytes), number of registered keywords (1 byte), keyword character code (1 byte), keyword string (32 bytes), file name (11 bytes) ), 0 reserving reservations (4 bytes), and individual data (32 bytes).

"File size" is the total size of the collective management file in blocks. "Number of blocks used" indicates the number of blocks using the set management file in the flash memory 22. "Connections" shows the number of connections when the collective management file is in reference to other files. "Date" indicates the date on which the administration file was created or updated. "Manufacturer / model code" indicates the manufacturer and model of the device which has written the collective management file on the memory card 20. "Initial registration directory number" is displayed as "0xffff" without description. "Registered keyword number" and "Keyword character code" are set to "0". The "file name" is used when performing management of the collective management file, and is not used in the memory card 20. FIG. “0 Reserve Reserve” is always set to 0 when executing duplicate entries. "Individual data" is all "0".

The file header shown in Fig. 8 includes standard identification data (8 bytes), file standard identification data (8 bytes), file ID (2 bytes), file translation (2 bytes), application creation date (8 bytes), and application update date (8 bytes). ), Maker / model code (4 bytes), update maker / model code (4 bytes), 0 reserving reservation (16 bytes), number of data entries (1 byte), number of tables (1 byte), character code (1 Bytes), a title string (128 bytes), and a reserved area (48 bytes).

"Standard identification data" indicates that the flash memory 22 stores the set management file according to a predetermined standard. "File standard identification data" indicates that the set management file is created according to the predetermined standard. "File ID" indicates the file type and is the same as the OS header part. "File translation" indicates the number of translations. The "Application Date" shows the date the collective management file was created, while the "Application Update Date" shows the date of the update. "Creating producer / model code" shows the author and model name that created the collective management file, while "update producer / model code" indicates the producer and model name that updated the collective management file. "Number of data entries" indicates the number of entries described later, and the number is three in the set management file. "Number of tables" indicates the number of data in the tail data area, where the number is indicated as "0". "Character code" denotes an input character by a predetermined code number, here "0xff". "Title string" indicates the title character, and all are "0xff".

The system entry list shown in Fig. 9 is a set of four sets of 12 bytes of data (entries) including a starting address (4 bytes), a data size (4 bytes), a data type ID (1 byte), and a reservation (3 bytes). It consists of. That is, the system entry list manages up to four entries. The set management file deletes the header, owns the bit map table, the address translation table and the connection information table, and thus contains information for managing these three tables.

In the example, for the bit map table, the "start address" indicates the address of the start position of the bit map table, and the "data size" indicates the number of blocks required for the bit map table. "Data type ID" describes "0x03" which indicates a bitmap table. The same configuration is used for the address conversion table and the connection information table.

The system configuration shown in Fig. 10 includes an empty block search position (2 bytes) and a reserved area (94 bytes). The "blank block search position" is provided for storing the search position of the blank block.

On the other hand, as shown in Fig. 11, the bit map table is used to manage the usage state of the flash memory 22 in units of blocks and includes distributed information and other head information for block management. Specifically, as shown in Fig. 11, the bitmap table is in the order of the block address in order of allow / disable (1 bit), reference flag (1 bit), system flag (1 bit), read only (1 bit), and display. Contains 8 bits of information consisting of (1 bit).

The "permit / disallow", "block flag" and "block final flag" are the same as the "block permit / disallow information", "block flag" and "block final flag" of the distributed management information. That is, "permit / disallow" indicates whether the block is available. The "block flag" indicates whether a block is unused, used as a leading block, used outside the leading block, and waiting for use and erasure. The "block final flag" indicates whether a block is contiguous or final in another file in one file. The "reference flag" indicates whether the additional information includes the referenced item. The "system flag" indicates whether the block is for general application or for system use. "Read only" indicates whether the block is read / writeable or read only. "Display" indicates whether the block is generally a block or a display block, and when it is for display, indicates a reference to the designated distributed management information.

The configuration of the address conversion table shown in Fig. 12 includes the physical addresses of each block in the order of the logical addresses of the blocks. That is, in the address conversion table, the block physical address is described so as to correspond to the logical address written in the distribution management information of each block. The conversion of the physical address and the logical address can be managed collectively. The physical address of an unused block is described as "0xffff".

The structure of the connection information table shown in FIG. 13 includes a connection address of distributed management information corresponding to each block as a "connection logical address" in order of block logical addresses. The "link logical address" is described as "0xffff" in the case of the last block, and "0x0000" in the case of the unused logical address. In addition, the logical address is dedicated for the root directory and the corresponding connection is prohibited.

In the memory card system 1 having the above configuration, when the CPU 17 of the host computer 10 is started up, as shown in FIG. 14, the CPU 17 stores the flash memory 22 of the memory card 20. ), The boot block information is read out and it is determined whether the set management information is included (step ST1 and step ST2).

In this process, the memory card 20 is also substantially started up, but it is explained here in accordance with the processing of the host computer 10.

If it is determined that the set management information is included, the CPU 17 reads the set management file from the flash memory 22 to the RAM 12 (step ST3). In addition, the CPU 17 judges whether the read out set management information is valid (step ST4). If it is determined that the set management information is valid, the CPU 17 builds a set management table in the RAM 12 to be used for memory management (step ST5).

Therefore, when the CPU 17 of the host computer 10 does not contain an error in the set management information of the memory card 20 at startup, the CPU 17 of the host computer 10 executes the set management information in order to execute management of data in the flash memory 22. Can read and execute high-speed processing within a short time after startup.

On the other hand, in step ST2, when the CPU 17 determines that the set management information is not included or in step ST4, when it is determined that the set management information is not valid, the CPU 17 ) Collects the distributed management information stored in the distributed management information area of each block so as to construct the collective management information by constructing the collective management information table in the storage means (step ST6).

Specifically, step ST6 executes the subroutine processing shown in FIG. The CPU 17 of the host computer 10 initializes the set management information storage memory table and registers the initial bad address of the boot block as unavailable in the bitmap table (step ST11), and blocks before the boot block are also unavailable. It registers (step ST12). Blocks after the boot block are checked in order, and it is determined whether all blocks have been checked (step ST13).

If it is determined that the checking of all blocks has not been completed, the CPU 17 reads out one block of distributed management information from the flash memory and stores it in the RAM 12 (step ST14). Thereafter, it is determined whether an error is included in the read distributed management information (step ST15). If an error is found, the error generated in the block is registered in the bitmap table (step ST16).

In addition, if it is determined that an error is not included in the read distributed management information, the CPU 17 registers a management flat of the distributed management information in the bitmap table and sets the physical address of the block corresponding to the address conversion table. The connection information corresponding to the logical address of the connection information table is described (step ST17). Therefore, the contents of the distribution management information of the block are described in the bitmap table, address conversion table and connection information table. In addition, the CPU 17 also checks whether a dual logical address or a connection address without a destination is included (step ST17).

After the CPU 17 describes the contents of the dispersion management information by the processing of the above steps ST13 to ST15 and ST17, the connection information table, address conversion table, and bitmap table are described. The process returns to step ST13 to describe the content of the information. When the reading of the distributed management information of all the blocks is finished, that is, when the checking of all the blocks is finished (step ST13), the CPU 17 performs a bitmap table, an address conversion table and an address to construct a set management file in the RAM 12. A header of 512 bytes is added to the data composed of the connection information table, and the construction of the group management information is finished. Therefore, when there is no set management information and the set management information is not valid, the set management file is constructed from the distributed management information.

At this time, the CPU 17 re-stores the set management information table in the RAM 12 as the set management information in the set management information area of the flash memory 22 for one period when the predetermined calculation processing such as power off is not executed. do.

Therefore, the CPU 17 can perform data management with the flash memory 22 based on the set management information stored in the flash RAM 12, and can easily read and write data at high speed. In addition, when there is no set management information or when the set management information is not valid, the distributed management information can be read out from the flash memory 22 to construct the set management information used for management. Therefore, even if the distributed management information is damaged, data management can be executed and reliability can be improved.

In addition, by constructing the bitmap table, the CPU 17 can recognize blocks that are not available based on the bitmap table of the group management information without retrieving the distributed management information of all the blocks, and execute writing of data at a higher speed. Can be. In addition, it is possible to know whether each block is a leading block, an intermediate block, or a block containing no data. This can also perform writing or reading of a file at high speed.

In addition, by building the address conversion table in the example, the CPU 17 can execute the physical address at a high speed at the logical address, and shorten the access time of the flash memory 22.

In addition, by establishing the connection information logical address, the CPU 17 can easily recognize the blocks constituting the file and can read a file composed of a plurality of blocks at high speed.

Next, a process of the CPU 17 for writing image data to the memory card 20 will be described with reference to FIG. 16.

When writing new data into the flash memory 22 or erasing data from the flash memory 22, the CPU 17 determines whether the set management information stored in the RAM 12 is valid (step ST21). Specifically, it is determined whether a part of the collective management information needs to be changed based on the distributed management information corresponding to the new data.

If it is determined that the set management information is valid, the CPU 17 executes a process of invalidating the set management information in the RAM 12 (step ST22) and writes new image data into the flash memory 22 or already stores it. The process of changing data such as erasing the image data thus obtained is executed (step ST23). If it is determined that the set management information is invalid in step ST21, a change process such as data writing is performed (step ST23).

The CPU 17 updates a part of the set management information in the RAM 12 based on the distributed management information of the block from which the image data and the like have been erased or the distributed management information corresponding to the newly written image data (step ST24).

At the end of the calculation processing such as immediately before the power-off, the CPU 17 writes the set management information stored in the RAM 12 into the set management information storage area of the flash memory 22 (step ST25).

That is, when data writing or erasing changes in the collective management information, the CPU 17 updates the collective management information based on the distributed management information such as the written data, and the collective management information always coincides with the distributed management information. Therefore, the reliability of the image data can be secured. In addition, by writing the group management information to the flash memory 22 at a suitable time, other calculation processing can be performed preferentially, thereby improving the overall calculation processing speed.

The memory card 20 of this embodiment has an external configuration as shown in Fig. 17 in the form of a flat thin card made of synthetic resin, for example.

In the memory card 20, for example, since an external terminal (not shown) is formed at one end 2a in the longitudinal axis direction, the memory card 20 is moved in the direction indicated by the arrow M in the drawing. (Not shown) can be attached.

In addition, in the memory card 20, for example, a lock cut-off portion can be formed on one side parallel to the mounting direction. , The electronic lock convex is engaged with the lock cut-off portion to prevent the memory card 20 from falling off from the electronic device.

That is, the memory card 20 of this embodiment has both the distributed management information stored in the distributed management storage area of the flash memory 22 and the set management information stored in the set management storage area as storage means. By managing these information in accordance with the data management method of the present invention, when detecting an error in the collective management information, the host computer 10 retrieves each distributed management information stored in the distributed management information storage area and retrieves the collective management information. It is possible to construct and use the collection management information easily even if the collection management information is damaged. Therefore, there is no risk that the reading and / or writing of the entire data is not executed. That is, the system is resistant to breakage.

In addition, when a conversion such as erasing and writing occurs in the distributed management information stored in the distributed management information storage area in the flash memory 22 of the memory card 20, the host computer 10 stores the stored information in the distributed management information storage area. The collective management information is updated based on the changed distributed management information, and the updated information is stored in the collective management information storage area in the flash memory 22 of the memory card 20. Therefore, there is no difference between the distributed management information and the collective management information in the memory card 20.

In addition, the next time the host computer 10 is started, the host computer retrieves the reconstruction and re-memory into the memory card as the collective management information, and the startup of the host computer 10 can be performed in a short time, and high-speed processing is possible. Do.

When the re-storage processing of the set management information is interrupted by the host computer 10 in the memory card 20, it is assumed that the set management information does not exist. In this case, when the next host computer 10 starts up, it is assumed that there is no set management information and processing is executed.

In addition, when the distributed management information is damaged in the memory card 20, data read and / or write execution corresponding to the block is impossible, but damage to the entire memory card 20 is minimized.

Although the above has described the case of the memory card for the serial interface, the data management apparatus according to the present invention can also be used for the parallel interface.

In the data management apparatus according to the present invention, the management means judges at the start whether the collective management information is valid, manages the data based on the collective management information when the collective management information is valid, and distributes when the collective management information is not valid. By managing the data based on the management information, it is possible to execute faster data processing when the data stored in the storage means is managed after startup. In addition, when the distributed management information is not valid, data can be managed by reading the distributed management information, so that reliability of data management can be ensured even when the set management information is damaged.

Further, the data management method according to the present invention determines whether the set management information is valid at startup, manages data based on the set management information when the set management information is valid, and based on distributed management information when the set management information is invalid. By managing the data, data can be managed at higher speed when data management is performed after startup. In addition, when the distributed management information is not valid, data can be managed by reading the distributed management information, so that reliability of data management can be ensured even when the set management information is damaged.

The storage medium according to the present invention stores arbitrary data in a data storage area configured for each predetermined storage unit, stores distributed management information for managing the data for each predetermined storage unit in the distributed management information area, and sets management information. By storing collective management information for collectively managing the data corresponding to the distributed management information in a storage area, information for managing data is distributed or integrated into one, so that data management is performed even if the collective management information is damaged. To ensure reliability.

1 is a block diagram showing the configuration of a memory card system according to the present invention.

2 is a block diagram showing a specific configuration of a memory card in the memory card system.

3 illustrates a block configuration of a flash memory in the memory card.

4 illustrates the contents of distributed management information for one page of blocks in the flash memory.

5 explains the contents of distributed management information used as additional information.

6 illustrates the configuration of a set management file.

7 illustrates the contents of the OS header in the set management file.

8 illustrates the contents of the file header in the header.

9 illustrates the contents of the system entry list in the header.

10 illustrates the contents of system information in the header.

11 illustrates the contents of a bit map table in the set management file.

Fig. 12 explains the contents of the address conversion table in the set management file.

13 explains the contents of the connection information table in the set management file.

14 is a flowchart for describing processing when the host computer is started up.

15 is a flowchart for describing processing when the host computer is started up.

Fig. 16 is a flowchart showing processing at the time of writing or erasing data.

17 shows an external perspective view according to the present invention.

* Explanation of symbols on the main parts of the drawings

1. Memory card system 10. Host computer

12.RAM 17.CPU

20. Memory Card 21. Control IC

22. Flash Memory

Claims (18)

Arbitrary data is stored in a data storage area configured for each predetermined storage unit, distributed management information for managing the data for each predetermined storage unit is stored in a distributed management information area, and the distributed management information is stored in a set management information storage area. Storage means for storing collective management information for collectively managing the corresponding data; It is determined at startup that the set management information is valid. If the set management information is determined to be valid, the data is managed based on the set management information. If it is determined that the set management information is invalid, the distributed management information. Management means for managing the data based on Data management apparatus comprising a. The method of claim 1, When the set management information is not valid, the management means constructs the set management information from the distributed management information stored in the distributed management information storage area, and writes the set management information into the storage means. Data management device. The method of claim 1, When a change occurs in a part of the set management information by writing new data in the data storage area of the storage means, the set management information is updated based on the distributed management information of the data written in the storage means, and the update is made. And write the collected set management information into the storage means. The method of claim 3, wherein And the management means writes the updated set management information into the storage means after the predetermined data ends, when executing the predetermined data processing. The method of claim 1, The storage means stores the distributed management information having a management flag for managing the storage state of the predetermined storage unit data and the set management information composed of the predetermined storage unit management flag; And the control means manages a data storage state of the data storage area based on the set management information. The method of claim 1, The storage means stores the distribution management information having a logical address of the predetermined storage unit and the set management information constituted by a set of physical addresses and logical addresses corresponding to the predetermined storage unit; The control means converts a physical address and a logical address of the predetermined storage unit based on the set management information, reads out data stored in the storage means, or writes data into the storage means. Management device. The storage means stores the distributed management information having connection information indicating that the predetermined storage unit is connected to another storage unit and the set management information constituted by a set of connection information items of the predetermined storage unit; And the control means reads out the data stored in the data storage area to which the predetermined storage unit is connected in the storage means based on the set management information. Storing arbitrary data in a data storage area configured for each predetermined storage unit; Storing distribution management information for managing the data for each predetermined storage unit in a distribution management information storage area; Storing the collective management information for collectively managing the data on the basis of the distributed management information in the collective management information area; Determining at start-up whether the set management information is valid; Managing the data based on the set management information when it is determined that the set management information is valid, or managing the data based on the distributed management information when it is determined that the set management information is invalid. Data management method comprising the step of. The method of claim 8, If the set management information is invalid, set management is made from the distributed management information stored in the distributed management information storage area, A data management method comprising storing the set management information. The method of claim 8, By writing new data in the data storage area, when a change occurs in a part of the set management information, the set management information is updated based on the distributed management information of the written data, And storing the updated set management information. The method of claim 9, In the case of processing predetermined data, after said predetermined data processing is finished, said collection management information is stored in said storage means. The method of claim 8, Storing the group management information composed of the management flag having the management flag for managing the storage state of the predetermined storage unit data and the management flag of the predetermined storage unit; And managing data storage of the data storage area based on the set management information. The method of claim 8 Storing distribution management information having a logical address corresponding to the logical address of the predetermined storage unit and a logical address corresponding to the physical address of the predetermined storage unit; Converting the logical address and the physical address of the predetermined unit based on the set management information, and reading out data stored in the storage means or storing new data. The method of claim 8, Storing distributed management information having connection information indicating that the predetermined storage unit is connected to another storage unit and the collection management information formed by collecting the connection information items of the predetermined storage unit; And reading out from the storage means stored in the data storage area to which the predetermined storage unit is connected, based on the set management information. And arbitrary data in a data storage area configured for each predetermined storage unit, and distributed management information for collectively managing the data on the basis of the distributed management information in the collective management information storage area. The method of claim 15, Distributed management information having a management flag for managing a storage state of the predetermined storage unit data; And a set management information comprising the set of the predetermined storage unit management flags. The method of claim 15, Distributed management information having a logical address of the predetermined storage unit; And a set management information comprising a set corresponding to a physical address and a logical address of the predetermined storage unit. The method of claim 15, Distributed management information having connection information indicating that a predetermined storage unit is connected to another predetermined storage unit; And a set management information comprising a set of connection information items of the predetermined storage unit.