JPH0764831A - Data storage device - Google Patents

Abstract

PURPOSE:To reduce a probability of data destruction by using small divisions (sectors) of a storage medium, where the write frequency is limited, on the average. CONSTITUTION:A storage medium part 11 where the frequency in data write is limited, a sector arrangement management part 14 which divides the storage medium part 11 into sectors to hold variable length data, and a sector state management part 15 which manages sectors in three states of the empty state, the busy state, and the used state are provided, and, sectors are only set to the used state at the time of data deletion, and these sectors are electrically erased when empty sectors are insufficient at the time of new data write. At the time of the occurrence of next file erase and write, respective unused memory retrieval priority addresses are moved in order to prevent only a specific memory place from being written and erased, thus extending the life of the memory.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor memory (hereinafter referred to as "EEPROM") that can be electrically written / erased, does not require a power supply dedicated to memory storage, and can retain the memory content by the element itself.
Called. ) For data storage and control information storage.

[0002]

2. Description of the Related Art In recent years, an EEPROM has come to be used as a storage device for a device using a microcomputer because it does not require a dedicated power supply for storing the stored data. Further, among these EEPROMs, a large-capacity storage element called a flash ROM is being developed and is being used in place of a mechanical storage device such as a magnetic disk.

In such a conventional magnetic disk device,
A small capacity block of a storage medium (magnetic disk) (hereinafter,
It is called a sector. ) Is used as the minimum unit for reading and writing, and the data is divided into a plurality of sectors to be written so that a large amount of continuous data can be stored. In addition, an operating system (hereinafter referred to as OS) that performs basic system management of the device adds and manages a data block name (hereinafter referred to as file name) necessary for managing stored data as a file. Therefore, the user can read and write in file units without being aware of the sectors.

FIG. 8 is a block diagram showing the structure of a conventional data storage device. In FIG. 8, reference numeral 80 is a data storage device, 81 is a storage medium unit composed of an electrically erasable semiconductor memory, and 82 is data transfer with an external control device 86 for instructing writing and reading of data. In order to temporarily store data, an input / output buffer unit 83, an entry management unit 83 that distinguishes a group of data by a file name, and a sector arrangement management unit 84 that records the order of used sectors, unused sectors, and used sectors. Reference numeral 85 denotes a control unit that receives a write or read instruction from the external control device 86 and instructs the data transfer process between the storage medium unit 81 and the external control device 86 via the input / output buffer unit 82. . Reference numeral 86 denotes the storage medium unit 8 via the input / output buffer unit 82.
This is an external control device that inputs and outputs data to and from 1, and instructs the control unit 85 to specify a file name, a total number of data, and write, read, and erase files.

When storing new data in the data storage area 1 of the storage medium unit 81, the entry management unit 83 manages a new file name and data length designated by the external control device 86 via the control unit 85. When the sector allocation management unit 84 finds an unused sector from the sector management table, the sector allocation management unit 84 receives the sector number and stores it in the file management table. Store as an entry point in. Further, when reading the data stored in the storage medium unit 81, the file name specified by the external control device 86 is searched from the file management table, and if found, the sector number indicated by the file name is used as the entry point. It is passed to the control unit 85 to prepare for reading data. If the specified file name cannot be found in the file management table, the control unit 85 is notified to that effect. When erasing data, the file name in the file management table is invalidated, the entry point to be erased is notified to the sector arrangement management unit 84, and a series of data is erased.

In response to an instruction from the entry management unit 83, the sector arrangement management unit 84 sequentially searches unused sectors from the sector number of the youngest (or oldest) sector when writing data, and when found, the sector number is found. Is notified to the entry management unit 83. Further, as the control unit 85 controls writing of a series of data to the storage medium unit 81, unused sectors of the sector management table are successively followed, and at the same time, a writable sector number is given to the control unit 85. Notice. On the other hand, when reading data, the control unit 85 is notified to read data from the sector number designated by the entry management unit 83, and the control unit 85 controls the external control device 86 to sequentially read data from the storage medium unit 81. While performing, the subsequent information in the sector management table is sequentially searched and continued until the end pointer value is encountered. Further, at the time of erasing data, a pointer value that means unused is set for all subsequent sectors in the sector management table from the sector number designated by the entry management unit 83 so that they can be treated as unused sectors where data can be written thereafter. To do.

Next, FIG. 9 showing a file management table for file management and sector management in the above-mentioned conventional example.
The sector management table will be described with reference to FIG. Here, it is assumed that the storage medium section 81 is divided into 256 bytes per sector. In FIG. 9, the file management table 90 is managed by the entry management unit 83 and stored in the data management area of the storage medium unit 81. 91
Is an entry number and is divided into areas 1 to m. Reference numeral 92 is a file name used to identify and use each data, 93 is a data size represented by the number of bytes, 94 is an entry pointer, and the head sector number in which the data is written is displayed. In this file management table 90, entry number 1 stores 600 bytes of data 1 from sector number 1, entry number 2 stores 16 bytes of data 2 from sector number 4, entry number 3 Indicates that 300 bytes of data 3 are stored starting from sector number 5.

In FIG. 10, the sector management table 10
0 is managed by the sector allocation management unit 84, and the storage medium unit 8
It is stored in the first data management area. Reference numeral 101 is a sector number, which is divided into storage areas 1 to n, 102 is a sector pointer, which indicates the usage status of each sector from sector number 1 to n, 0 is empty, and FF is The number other than 0 and FF indicates the sector number succeeding the sector number of itself. The file name is data 1 in the file management table 90.
Data of 600 bytes, three sectors of 256 bytes are required, and sector numbers 1, 2, and 3 are used. Similarly, since the data 2 is 16 bytes, the sector is 1
The number of sectors is 4 and the sector number 4 is used. Since the data 3 is 300 bytes, two sectors are required, and the sector numbers 5 and 6 are used. Since sector number 7 and later are empty, 0 is assigned.

When writing data to the storage medium section 81,
The file management table 90 is searched from the entry 1 to the entry m by the entry management unit 83, and the search operation is stopped at the entry m. In the example of FIG. 9, the entry used for the next data write is entry 4, the entry management unit 83 writes the file name in the file name of entry 4, and the number of data bytes in the data length column 93. . Subsequently, the sector allocation management unit 84 searches the sector pointer 102 of the sector management table 100 from sector 1 to sector n, and when an unused sector is found, the entry management unit 83 is notified of that fact.
The sector address is stored in the entry point 94 in the file management table 90. In this example, the address of sector 7 is stored in entry point 94 of entry number 4. Subsequently, the data written in the data storage area of the storage medium unit 81 will be stored in the sector allocation management unit 84 after that.
Control of the sector management table 100, the next empty sector of the sector pointer 102 (sector 8 in the example of FIG. 10).
Sector pointer 102 of the previous sector 7 so that
And the same processing is performed as long as the data continues. Further, the data erasure does not erase all the data in the erased sector of the storage medium unit 81, but the file name 92 of the file management table 90 and the sector management table 1
The data remaining in the storage medium unit 81 is rewritten for the first time when the next data is written.

[0010]

However, in the above-mentioned conventional configuration, the state of each sector is managed only when it is empty or in use (successor sector number or FF).
When writing or erasing frequency of a sector cannot be managed and new data is to be written, the sector management table of the sector allocation management unit will search for an empty sector in ascending or descending order of sector number, depending on which method is used. If you use it as an external storage device that uses a semiconductor memory that has a limited number of erasures, the frequency of writing or erasing the lower or higher number increases, and the possibility of data errors increases and the life of the storage device decreases. Had the problem.

An object of the present invention is to solve the above-mentioned conventional problems, and an object thereof is to provide a data storage device capable of improving the reliability of data and extending the life of the storage device.

[0012]

In order to achieve the above-mentioned object, the present invention prepares a sector state of being empty, a state of being used in addition to a state of being used, and erasing by an electric signal at the time of erasing data. Do not write the data immediately, but only if the external controller issues a new data write request and the number of empty sectors is insufficient, perform the erase operation on the used sectors to return to the empty state and then write the data. It is the one.

The present invention is also one in which the search start point for searching the file management table and the sector management table is sequentially moved every time data is erased.

[0014]

According to the present invention, due to the above configuration, the file is not ready to be used only by being erased, and the erase operation is performed by the electric signal only when almost all the sectors are in use or once used. Therefore, the search start points of the file management table and the sector management table are moved sequentially, so that the erase operation count is averaged and only a specific sector does not exceed the guaranteed erase count. The storage device life can be extended.

[0015]

【Example】

(Embodiment 1) An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a data storage device according to the first embodiment of the present invention. In FIG.
Reference numeral 10 is a data storage device according to the present embodiment, 11 is a storage medium unit composed of an electrically erasable semiconductor memory, 1
Reference numeral 2 denotes an input / output buffer unit that temporarily stores data to be transferred to and from the external control device 17 that instructs writing and reading of data, 13 an entry management unit that distinguishes a group of data by a file name, and 14 A sector arrangement management unit that records the order of used sectors, unused sectors and used sectors, 15 a sector state management unit that manages the state of sectors,
Reference numeral 16 denotes a storage medium unit 1 via the input / output buffer unit 12 in response to a writing or reading instruction from the external control device 17.
1 is a control unit for instructing a data transfer process between the external control device 1 and the external control device 17.

File management and sector management in the first embodiment will be described with reference to FIG. 2 showing a file management table and FIG. 3 showing a sector management table. Here, it is assumed that the storage medium unit 11 is divided into 256 bytes per sector. In FIG. 2A, a file management table 20 managed by the entry management unit 13 is stored in the data management area of the storage medium unit 11. 11 is an entry number, which is divided into areas 1 to m. Reference numeral 12 is a file name used as an identification of each data, 13 is a data size represented by the number of bytes, 14 is an entry pointer, and the head sector number in which the data is written is displayed. In the file management table 20, entry number 1 stores 600 bytes of data 1 from sector number 1, entry number 2 stores 16 bytes of data 2 from sector number 4, entry number 3
Indicates that 300 bytes of data 3 are stored starting from sector number 5. FIG. 2B shows data 2 of entry number 2 in this file management table 20.
Is erased, and 700 bytes of data 4 is newly stored therein.

In FIG. 3A, reference numeral 30 denotes a sector management table managed by the sector arrangement management unit 14, which is stored in the data management area of the storage medium unit 11. Reference numeral 31 is a sector number, which is divided into storage areas 1 to n, 32 is a sector pointer, which indicates the usage status of each sector from sector numbers 1 to n, 0 is empty, and FF is The number other than 0 and FF indicates the sector number succeeding the sector number of itself. FIG. 3B shows that the sector number 4 becomes empty in the sector management table 30, and FIG. 3C shows a state in which data is written in the sector number 4 and the sector number 7 follows. ing.

In FIG. 4A, reference numeral 40 denotes a sector status management table managed by the sector status management unit 15, which is stored in the data management area of the storage medium unit 11. 41
Is a sector number, and 42 is a sector use state. As for the sector state, 0 indicates empty, 1 indicates in use, and 2 indicates used. FIG. 4B shows that the sector number 4 has become used 2, and FIG. 4C shows that the sector numbers 4, 7, and 8 have become used 1. The used 2 state is an empty state in terms of management, but it is necessary to perform an erasing operation by an electric signal before newly writing.

Next, the operation of deleting the data 2 and newly writing the data 4 in the storage medium section 11 will be described. First, the data 2 is deleted from the file name in the file management table 20 of the entry management unit 13 of FIG. 2A, and the sector number 4 in the entry pointer 24 is notified to the sector arrangement management unit 14 and the sector state management unit 15. Upon receipt of this notification, the sector arrangement management unit 14 rewrites the sector pointer FF of the sector number 4 from the sector management table 30 of FIG. 3A to 0, which indicates an empty state, as shown in FIG. 3B. In addition, the sector state management unit 15 is shown in FIG.
The in-use 1 of the sector number 4 in the sector state management table 40 of (a) is rewritten to the used 2 as shown in FIG. 4 (b). In this way, the data 2 and the series of data related thereto are erased.

Next, the operation of writing the data 4 of 700 bytes after the erased data 2 will be described. First, the file name and the data length of the data 4 are sent from the external control device 17 via the control unit 16 to the entry management unit 13, and the data 2 of the file management table 20 as shown in FIG.
The data 4 is registered in the file name 22 of the entry number 2 in which the error occurred, and 700 bytes is registered in the data length 23.
Next, the entry management unit 13 requests the sector arrangement management unit 14 and the sector state management unit 15 to search for unused sectors. Now, as shown in FIG. 3B, the sector management table 3
0, only the sector numbers 4, 7, and 8 are unused sectors in the empty state 0 between sectors 1 to n.
As shown in (b), in the sector state management table 40, the only unused sectors in the empty state 0 are sector numbers 7 and 8.
Since data 4 is 700 bytes, three sectors are required. Therefore, the sector state management table 4 in FIG.
Sector number 4 in 0 is empty from the state of used 2
The unused sector numbers 4, 7, and 8 are secured by rewriting to the state of (3), and the entry management unit 13 is notified. As a result, as shown in FIG. 2B, the entry management unit 13 registers 4 as the entry pointer of the entry number 2 and the control unit 16 stores the data 4 in the data storage area of the storage medium unit 11.
When the sector number 4 is full, the data is recorded in sector number 7 and then sector number 8. During this period, as shown in FIG. 3C, the sector arrangement management unit 14 writes the subsequent sector number 7 to the sector pointer 32 of sector number 4, and then writes the subsequent sector number 8 to the sector pointer 32 of sector number 7. FF with no trailing in the sector pointer 32 of sector number 8
Write. In addition, the sector state management unit 15 is configured as shown in FIG.
As shown in (c), all sector numbers 4, 7, and 8 are in use 1
Rewrite to the state of.

As described above, according to the first embodiment described above, when the file is erased, the sectors used by the file are not immediately set to the free state, but the sector state management unit 15 keeps them as they are. In this case, only when almost all the sectors are in use and the number of free sectors is insufficient, the erase operation is performed on the used sectors to return to the free state. Since the write operation is performed, the erase operation is averaged and it is possible to prevent the limit of the guaranteed erase count in a specific sector from being exceeded.

(Second Embodiment) Next, a second embodiment of the present invention will be described. FIG. 5 shows the configuration of a data storage device according to the second embodiment of the present invention. In FIG. 5, reference numeral 50 denotes a data storage device according to the present embodiment, 51 denotes a storage medium unit formed of an electrically erasable semiconductor memory, and 52 denotes an external control device 57 for instructing writing and reading of data. An input / output buffer section for temporarily storing the data to be transferred, 53 an entry management section for distinguishing a group of data by a file name, 54 a sector arrangement management for recording the order of used sectors and unused sectors The unit 55 sequentially moves the search start points when searching the unused areas of the file management table in the entry management unit 53 and the sector management table in the sector arrangement management unit 54 according to an instruction from the control unit 56 when erasing data. Search order management unit, 56 is the external control device 5
In response to the write / read instruction from the storage medium unit 7, the storage medium unit 51 and the external control device 57 are passed through the input / output buffer unit 52.
And a control unit for instructing a data transfer process between and.

File management and sector management in the second embodiment will be described with reference to FIGS. 6 and 7. FIG. 6A shows a file management table managed by the entry management unit 53, which is stored in the data management area of the storage medium unit 51. File management table 6
0 is the entry number 61, file name 62, data length 6
3 and an entry pointer 64. Figure 6
(B) shows a file search order management table managed by the search order management unit 55, which is stored in the data management area of the storage medium unit 51. File search order management table 6
5 corresponds to the file management table 60, and is composed of an entry number 66 and a search order 67.

FIG. 7A shows a sector management table managed by the sector arrangement management unit 54, which is stored in the data management area of the storage medium unit 51. Sector management table 70
Is composed of a sector number 71 and a sector pointer 72. FIG. 7B shows a sector search order management table managed by the search order management unit 55, and the storage medium unit 51.
Are stored in the data management area of. The sector search order management table 73 corresponds to the sector management table 70,
It is composed of a sector number 74 and a search order 75.

The file search order management table 65 managed by the search order management unit 55 is composed of 1 bit (may be a plurality of bits), and the search order management unit 55 sets 1 to the next entry number for each file deletion. It is controlled to be added. In the example of FIG. 6B, the entry number 8 which is the point where the bit 1 changes from 0 in the search order 67
Is recognized as the search start point P, the bit of the entry number 8 becomes 1 at the next file deletion, and the next entry number 9 becomes the search start point P. File management table 60
The search for an unused entry number in 8 starts from the entry number 8 of the search start point P by the entry management unit 53,
The process proceeds to the entry number m, and then from the entry number 1 to the entry number 7 immediately before the search start point P.

Similarly, the sector search order management table 73 managed by the search order management unit 55 is also composed of 1 bit (may be a plurality of bits).
It is controlled so that 1 is added to the next sector number each time a file is erased. In the example of FIG. 7B, the sector number 8 which is the point at which the bit 1 changes from 0 to 0 in the search order 75
Is recognized as the search start point P, the bit of the sector number 8 becomes 1 at the next file erasure, and the next entry number 9 becomes the search start point P. The sector allocation management unit 5 searches for an unused entry number in the sector management table 70.
4 starts from the sector number 8 of the search start point P, proceeds to the sector number n, and then newly starts from the sector number 1 to the sector number 7 immediately before the search start point P.

The search start point P may be moved in either descending order or ascending order. In the descending order, when the oldest number is reached, the search is resumed and the descending order starts from the youngest number, and in the ascending order. , When the youngest number is reached, it will return to the original number and will be in ascending order from the oldest number.

As described above, according to the second embodiment, when the external control device 57 instructs the control unit 56 to delete a file, the entry management unit 53 and the sector arrangement management unit 54 respectively manage the files. When searching the table 60 and the sector management table 70, referring to the file search order management table 65 and the sector search order management table 73, the search is started from the position of the search start point P, and there is an instruction to erase the next file. In this case, since the search start point P is sequentially shifted to the next position, the access frequency of the entire memory element for writing and erasing can be made uniform, and only the memory element at the specific address is partially written. It is possible to prevent the erasure limit from being exceeded.

[0029]

According to the present invention, as is apparent from the above embodiment, the state of the sector is set to empty, used, and used, and the erasure is immediately performed by the electric signal when erasing the data. Instead, only when there is a new data write request from the external control device and the number of empty sectors is insufficient, the erase operation is performed on the used sectors to return to the empty state, and then the data is written. Therefore, the erase operation is averaged, and the guaranteed erase count does not exceed the specific sector, so that the reliability of data can be improved and the life of the storage device can be extended.

Further, according to the present invention, at the time of erasing data, the search start points for searching the unused areas of the file management table in the entry management unit and the sector management table in the sector arrangement management unit are sequentially moved. , It is possible to make the access frequency of the entire memory device for writing and erasing uniform, and only the memory device at a specific address does not partially exceed the limit of the number of times of writing and erasing, thereby improving the reliability of data, The life as a storage device can be extended.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a data storage device according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram showing an operation of a file management table of an entry management unit in the first exemplary embodiment of the present invention.

FIG. 3 is a schematic diagram showing an operation of a sector management table of a sector arrangement management unit in the first exemplary embodiment of the present invention.

FIG. 4 is a schematic diagram showing an operation of a sector status management table of a sector status management unit in the first embodiment of the present invention.

FIG. 5 is a block diagram showing a configuration of a data storage device according to a second embodiment of the present invention.

FIG. 6 is a schematic diagram showing operations of a file management table of an entry management unit and a file search order management table of a search order management unit according to the second embodiment of the present invention.

FIG. 7 is a schematic diagram showing operations of a sector management table of a sector arrangement management unit and a sector search order management table of a search order management unit according to the second embodiment of the present invention.

FIG. 8 is a block diagram showing a configuration of a data storage device in a conventional example.

FIG. 9 is a schematic diagram showing a file management table of an entry management unit in a conventional example.

FIG. 10 is a schematic diagram showing a sector management table of a sector arrangement management unit in a conventional example.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 data storage device 11 storage medium unit 12 input / output buffer unit 13 entry management unit 14 sector allocation management unit 15 sector state management unit 16 control unit 17 external control device 50 data storage device 51 storage medium unit 52 input / output buffer unit 53 entry management Part 54 Sector placement management part 55 Search order management part 56 Control part 57 External control device

Claims (2)

[Claims] 1. A storage medium unit composed of a non-volatile semiconductor memory in which data can be erased by an electric signal, and the storage medium unit is divided into sectors which are storage sections in which data can be collectively erased, and the sector is the head of data. , A sector arrangement management unit that holds variable-length data by recording the final order on the way, and an entry management unit that stores a plurality of variable-length data by holding a file name, data size, and the start number of the sector Then, the sector is electrically erased and data can be written at any time, an empty state in which the data is held, and the written data becomes unnecessary but before the electrical erasure is performed. A sector status management unit that manages in three used states that cannot be newly written, and an I / O buffer unit that is used as a write / read work area And a control unit that uses the input / output buffer unit to read / write data from / to the storage medium unit. 2. A storage medium section composed of a non-volatile semiconductor memory in which data can be erased by an electric signal, and the storage medium section is divided into sectors which are storage sections in which data can be collectively erased, and the sector is the head of data. , A sector arrangement management unit that holds variable-length data by recording the final order on the way, and an entry management unit that stores a plurality of variable-length data by holding a file name, data size, and the start number of the sector An input / output buffer unit used as a writing / reading work area; a control unit for reading / writing data from / to the storage medium unit using the input / output buffer unit; Unused areas of the file management table in the entry management section and the sector management table in the sector allocation management section A data storage device including a search order management unit that moves a search start point when searching a region to the next position.