JP2009288924A - File access system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that access can be extremely delayed since there is a need to cope with a cost-down request in the conventional file system. <P>SOLUTION: This file access system includes a storage part and a control part. The storage part stores a table in which a first column for storing cluster information showing the position of a region where directory entry of a folder or a file belonging to a folder is stored for each folder and a second column for storing the file number of the file in which the directory entry is stored at a position shown by the cluster information stored in the first column, are defined. When the file number of the file stored in the storage medium is given, the control part refers to the table of the storage part and acquires the start position of retrieval. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a file access system.

  The FAT (File Allocation Table) is a table indicating the allocation state of data areas secured in the storage medium, and is managed in units called clusters. When writing the contents of a file to the data area of the storage medium, if the data capacity of the file exceeds one cluster, writing to the next cluster is performed. At this time, in the FAT, information indicating the next cluster number is stored in order to indicate that the file contents are also written in the next cluster. The FAT format is adopted in many storage media such as a hard disk and a removable disk.

  FIG. 1 is a diagram for explaining a general data structure of a storage medium formatted in the FAT format. In FIG. 1, a partition boot sector is an area for storing data for booting such as a partition parameter value. [FAT1] is an area for storing the first FAT, and [FAT2] is an area for storing the second FAT for improving reliability. [FAT1] and [FAT2] store in which position (which cluster number) the contents of the folder or file are in the data area. The root directory entry is an area for storing the contents of the root folder. The root directory entry stores start position information and names of folders and files belonging to the root folder. The data area is an area for storing the contents of folders and files other than the root folder.

  FIG. 2 is an explanatory diagram of the data structure of the directory entry. In FAT, when one folder or file is created, one directory entry is created. As shown in FIG. 2, the directory entry has a size of 32 bytes and includes a file name (folder name), an extension, an attribute, a start cluster number, a file size, and the like. The start cluster number indicates where in the data area the contents of the folder or file are stored.

  When reading the contents of a certain folder or file, the directory entries belonging to the root folder are searched in order. By referring to the directory entry, the start position of the target folder or file can be determined. When the start position is determined, [FAT1] and [FAT2] are searched using the start position. With [FAT1] and [FAT2], it is possible to confirm the cluster connection in which the contents of the target folder or file are stored in the data area. When the cluster connection is confirmed, the contents of the folder or file can be acquired by reading the data in the order of the connection.

  3 shows an example of the directory structure of the storage medium, FIG. 4 shows the directory entry of the folder or file belonging to the root folder of FIG. 3, and FIG. 5 shows the directory entry of the folder or file belonging to the SUB_DIR1 folder of FIG. , Respectively. In this example, as shown in FIGS. 3 and 4, [FILE1], [SUB_DIR1], [SUB_DIR2],..., [FILE5] belong to the root folder. As shown in FIGS. 3 and 5, [FILE_6], [SUB_DIR11], [SUB_DIR12],..., [FILE200],..., [FILE255] belong to the [SUB_DIR1] folder. . Referring to the root directory entry in FIG. 4, since the start cluster of the [SUB_DIR1] folder is “0x0008”, the [SUB_DIR1] directory entry in FIG. 5 is stored from the position “0x0008”. Recognize.

  Japanese Patent Application Laid-Open No. 2002-163136 (see Patent Document 1) describes an invention of a file system that can access a FAT-type storage medium at high speed by creating a cluster link table in a storage means. In the invention of this file system, the cluster link table is created for some folders or files, so the capacity of the storage means can be reduced. Further, since the cluster connection can be confirmed by referring to the cluster link table stored in the storage means, it is possible to access folders and files at high speed.

JP 2002-163136 A

  In the invention described in Japanese Patent Laid-Open No. 2002-163136, if a folder or file clustering table is stored in the storage means, the folder or file can be accessed at high speed, but if it is not stored, It is necessary to perform a search again from the root directory entry of the storage medium. Therefore, in some cases, there arises a problem that access is extremely slow.

  Further, due to the recent cost reduction request, it is necessary to reduce the capacity of the storage means. If the capacity of the storage means is reduced, the area for storing the cluster link table is inevitably reduced, and the number of folders and files that can create the cluster link table is reduced. Therefore, there is a possibility that the problem of extremely slow access frequently occurs due to the response to the cost reduction request.

  [Means for Solving the Problems] will be described below using the numbers and symbols used in [Best Mode for Carrying Out the Invention]. These numbers and symbols are added in parentheses in order to clarify the correspondence between the description of [Claims] and [Best Mode for Carrying Out the Invention]. However, these numbers and symbols should not be used for the interpretation of the technical scope of the invention described in [Claims].

  The file access system (15) according to one aspect of the present invention includes a storage unit (11) and a control unit (12). The storage unit (11) stores, for each storage medium (10) formatted in the FAT format, cluster information indicating the position of an area in which a directory entry of a folder or a file belonging to the folder is stored. One column and a file number assigned in the order of folders and in the order of arrangement of directory entries in the folder, and the directory at the position indicated by the cluster information stored in the first column A first table storage area (111) is defined in which a second field for storing a file number of a file in which an entry is stored is defined. When the control unit (12) is given the file number of the file stored in the storage medium (10), the control unit (12) refers to the first table and, from the second column, the first neighborhood closer to the given file number. Search the file number, obtain the directory entry position of the file having the first neighboring file number from the first column, and search the directory entry of the file having the given file number using this position as the start position To do.

  The storage unit (11) further stores in the storage area (111) a second table in which a file number is associated with cluster information indicating a position where a directory entry of a file having the file number is stored. be able to. When the file number of the file stored in the storage medium (10) is given, the control unit (12) refers to the second table and searches for the second neighboring file number close to the given file number, The position of the directory entry of the file having the second neighboring file number is obtained, and the directory entry of the file having the given file number can be searched using this position as a start position. Further, when searching for the first and second neighboring file numbers, the control unit (12) compares the first and second neighboring file numbers, selects one according to a predetermined criterion, and selects the first or second neighboring file number. The location of the directory entry of the file having the first or second neighboring file number can be obtained from the second table.

  According to the present invention, folders and files can be accessed at high speed by using a storage area table. Further, the storage area for storing the table can be reduced in capacity.

  One of the best modes for carrying out the present invention will be described in detail with reference to the drawings. FIG. 6 shows a configuration explanatory diagram of the first embodiment. In FIG. 6, the file access system 15 includes a storage unit 11 and a control unit 12, and accesses the storage medium 10 formatted in the FAT format. The control unit 12 reads necessary information from the storage medium 10 and creates a predetermined start position table. Then, the start position table is stored in the start position table area 111 of the storage unit 11. Further, the control unit 12 refers to the start position table stored in the start position table area 111, searches for the directory entry of the designated file, and accesses the file at high speed. In the storage unit 11, a work area 113 is secured together with the start position table area 111.

  In the first embodiment, a folder information start position table and a file information start position table are stored in the start position table area 111 of the storage unit 11. FIG. 7 is an explanatory diagram of the data structure of the folder information start position table, and FIG. 8 is an explanatory diagram of the data structure of the file information start position table.

  The folder information start position table in the first embodiment will be described. As shown in FIG. 7, the folder information start position table has a folder number column, a start cluster column, and a start file number column. The folder number column stores a number uniquely assigned to each folder, and each folder can be uniquely determined by the folder number. The column of the start cluster is a column for storing position information as a starting point for starting the search. In the first embodiment, the start cluster of the area in which the directory entry of the folder or file belonging to the folder is stored A number is stored. The start file number column stores the file number of the file that is the starting point of the search. In the first embodiment, the start file number column has the smallest file number among the valid files belonging to the folder. This field stores the file number of the file.

  In the first embodiment, file numbers are assigned in the order of folder numbers. For example, the file number of the valid file belonging to the folder with the folder number 1 is smaller than the file number of the valid file belonging to the folder with the folder number 2. In each folder, file numbers are assigned in the order of directory entries. For example, the directory entry position of the file with the file number 6 is ahead of the directory entry position of the file with the file number 7.

  The folder information start position table in FIG. 7 has, as options, a higher folder number field, a valid file number field, and a folder information position field. The upper folder number column is a column for storing the folder number of the upper folder. However, zero is stored in the case of the root folder. The column for the number of valid files is a column for storing the number of valid files. The valid file refers to a specific type of file designated in advance among files stored in the storage medium 10, for example, a file whose file extension is mp3 or wma. The folder information location column is a column for storing information indicating the location where the directory entry of the folder is stored.

  A method for creating the folder information start position table in FIG. 7 will be described. The control unit 12 reads necessary information from the storage medium 10, calculates the number of valid files for each folder, and assigns a unique file number to each valid file. In the example of FIG. 7, consecutive numbers are assigned in the order of folder numbers and in the order of the directory entries in each folder. Then, the control unit 12 stores the file number of the valid file having the smallest file number among the valid files belonging to the folder in the start file number column. In this way, the folder information start position table of FIG. 7 can be created.

  The file information start position table in the first embodiment will be described. As shown in FIG. 8, the file information start position table has a column for belonging folder number and a field for file information position. The column of belonging folder number is a column for storing the folder number of the folder to which the file belongs. The file information position column is a column for storing the directory entry position of the file.

  For example, focus on a file with a file number of 500. Referring to FIG. 7, the file with the file number 500 belongs to the folder with the folder number M, and its start cluster is m. Therefore, in the second table entry of FIG. 8, M is stored as the belonging folder number, and m is stored as the cluster number of the file position information.

  The file information start position table in the first embodiment stores file information while keeping an interval between file numbers, and does not have the number of table entries corresponding to the number of all valid files. In FIG. 8, the file information is stored with the interval between the file numbers being a fixed number of 250. Further, as illustrated, the position of each table entry is associated with a file number. For example, the first table entry is associated with a file number of 250, and the second table entry is associated with a file number of 500.

  A method of using the folder information start position table of FIG. 7 and the file information start position table of FIG. 8 will be described. When the file number is designated, the control unit 12 searches for the directory entry of the file having the file number. At this time, it is determined which of the folder information start position table and the file information start position table is used to start the search faster.

  For example, a case where the designated file number is 255 will be described. When a table entry with a start file number of 255 or less is searched from the folder information start position table of FIG. 7, a table entry with a folder number of 2 is searched. Therefore, in the folder information start position table of FIG. On the other hand, when the table entry associated with the file whose file number is 255 or less is retrieved from the file information start position table of FIG. 8, the table entry associated with the file number of 250 is retrieved. Therefore, in the file information start position table of FIG. 255 is closer to 250 than 6 is. Therefore, the file information start position table in FIG. 8 is selected. The control unit 12 refers to the table entry of the file whose file number is 250 from the file information start position table of FIG. 8, and acquires the cluster number “0x0008” and the offset value “#Q”. Then, the search is started from here, and the directory entry of the file whose file number is 255 is found.

  As another example, a case where the designated file number is 257 will be described. When a table entry whose start file number is 257 or less is searched from the folder information start position table of FIG. 7, a table entry whose folder number is 3 is searched. Therefore, in the folder information start position table of FIG. On the other hand, when the table entry associated with the file whose file number is 257 or less is retrieved from the file information start position table of FIG. 8, the table entry associated with the file number of 250 is retrieved. Therefore, in the file information start position table of FIG. 257 is closer to 256 than 250. Therefore, the file information start position table in FIG. 7 is selected. The control unit 12 refers to the table entry with the folder number 3 from the file information start position table of FIG. 7, and acquires the start cluster “0x000E”. Then, the search is started from “0x000E” as a starting point, and the directory entry of the file whose file number is 257 is found.

  The effect of the first embodiment will be described. The first effect is faster access. Each processing time is 500 ms from the root directory entry to analyzing and storing the folder position information, 250 ms from analyzing and storing the file information in the folder, and acquiring the actual file head position from the file information table. It takes 5 ms. At this time, in the technique of Patent Document 1, it takes 5 ms when there is information in the cluster link table, and 755 ms when there is no information. However, in the first embodiment, when there is information in the table, it is 5 ms, and when there is no information, it is 255 ms, so the maximum analysis time is shortened.

  The second effect is a capacity reduction effect of the storage unit. For example, assume that there are 700 folders and 65535 files. Each table entry of the folder information start position table requires 12 bytes (the folder information start position table assumed as a comparison reference has no start file number column. In the first embodiment, In the folder information start position table, since the column of the start file number is added to the folder information start position table assumed as a comparison, 14 bytes are required for each table entry.) Assume that 4 bytes are required for each table entry. At this time, assuming a comparative example in which table entries for all folders and all files are provided in two tables, the capacity of the storage unit 11 excluding the work area 113 is 12 × 700 = 8400 bytes in the folder information start position table. The information start position table requires 4 × 65535 = 262140 bytes, and a total of 270540 bytes (about 264 Kbytes). On the other hand, in the first embodiment, 14 × 700 = 9800 bytes are required for the folder information start position table, and 65535 ÷ 250 × 4 = 262 × 6 = 1572 bytes are required for the file information start position table, and 11372 bytes in total. (About 11 Kbytes). Thus, the capacity of the storage unit is considerably low.

  Next, another best mode for carrying out the present invention will be described. The configuration of the second embodiment is the same as that of the first embodiment, and will be described with reference to FIG. However, in the second embodiment, a folder information start position table similar to that in the first embodiment and a file information start position table different from that in the first embodiment are stored in the start position table area 111. To do.

  FIG. 9 is an explanatory diagram of the data structure of the file information start position table in the second embodiment. Like the file information start position table in the first embodiment, the file information start position table in the second embodiment has a column of the belonging folder number and a file information position column. The method of providing table entries is different, and the number of table entries is considerably reduced.

  A method for creating the file information start position table in FIG. 9 will be described. In the second embodiment, the control unit 12 is given a predetermined threshold for the upper limit of the number of valid files in a folder. Then, a folder having the number of valid files exceeding the threshold is extracted. For example, assume that the predetermined threshold is 250. Referring to the folder information start position table in FIG. 7, the folder having the number of valid files 501, that is, the folder with the folder number N is extracted.

  Following the folder extraction, the control unit 12 performs folder division. For example, when dividing using the threshold value as it is, the folder with the folder number N is first divided at 915 + 250 = 1165. The control unit 12 adds the file information having the file number 1165 to the first table entry of the file information start position table of FIG. Further, the folder with the folder number N is divided by 915 + 250 + 250 = 1415. The control unit 12 adds the file information whose file number is 1415 to the second table entry of the file information start position table of FIG. In this way, a file information start position table as shown in FIG. 9 can be created.

  A method of using the folder information start position table of FIG. 7 and the file information start position table of FIG. 9 in the second embodiment will be described. When the file number is designated, the control unit 12 searches for the directory entry of the file having the file number. At this time, first, the control unit 12 determines which one of the folder information start position table and the file information start position table is used to start the search.

  For example, a case where the designated file number is 255 will be described. In the folder information start position table of FIG. 7, the file number for starting the search is 6, and in the file information start position table of FIG. 9, there is no file to start the search. Therefore, the control unit 12 selects the folder information start position table of FIG. 7 and starts searching from the file with the file number 6 from the start cluster “0x0008” of the folder with the folder number 2.

  As another example, a case where the designated file number is 1200 will be described. In the folder information start position table of FIG. 7, the file number for starting the search is 915, and in the file information start position table of FIG. 9, the file number for starting the search is 1165. Therefore, the control unit 12 selects the file information start position table of FIG. Then, the search is started from the position where the directory entry of the file having the file number 1165 exists, the cluster n, and the offset #Q.

  In the second embodiment, the capacity of the file information start position table can be further reduced. Moreover, since the folder is divided by a predetermined threshold value, the directory entry of the file having the designated file number can be found by searching for the number of directory entries within the range not exceeding the threshold value. Accordingly, it is possible to avoid an extreme decrease in the access time of folders and files.

  As other embodiments, only the folder information start position table of FIG. 7 is stored in the start position table area 111, and the file information start of FIG. 8 is started without storing the folder information start position table of FIG. An embodiment that stores the position table and the file information start position table of FIG. 9 can also be realized.

General data structure of storage media formatted in FAT format Data structure of directory entry The figure which shows the example of the directory structure of a storage medium The figure which shows the directory entry of the folder or file which belongs to the root folder of FIG. The figure which shows the directory entry of the folder or file which belongs to the SUB_DIR1 folder of FIG. Configuration explanatory diagram of the first and second embodiments Explanatory drawing of the data structure of the folder information start position table in the first and second embodiments Data structure explanatory diagram of the file information start position table in the first embodiment Explanatory drawing of the data structure of the file information start position table in the second embodiment

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Storage medium 11 Storage part 12 Control part 15 File access system 111 Start position table area 113 Work area

Claims (5)

For a storage medium formatted in the FAT (File Allocation Table) format, for each folder, a first column for storing cluster information indicating a location of a folder or a directory entry of a file belonging to the folder; This is a column for storing file numbers assigned in the order of folders and in the order of arrangement of directory entries in the folder, and the directory entries are stored at the positions indicated by the cluster information stored in the first column. A storage unit comprising a storage area of a first table in which a second field for storing a file number of a file being defined is defined;
Given the file number of the file stored in the storage medium, refer to the first table, search the first column for the first neighboring file number close to the given file number from the second column, A control unit that obtains the position of the directory entry of the file having the first neighboring file number from the first column, and searches for the directory entry of the file having the given file number using the position as a start position; A file access system comprising: The storage unit
In the storage area, further storing a second table in which a file number is associated with cluster information indicating a position where a directory entry of a file having the file number is stored;
The controller is
When the file number of the file stored in the storage medium is given, the second neighbor file number close to the given file number is searched with reference to the second table, and the second neighbor file number is The file access system according to claim 1, wherein a directory entry position of a file having a given file number is retrieved using the position of the directory entry of the file having the given file number as a start position. The controller is
When the first and second neighboring file numbers are searched, the first and second neighboring file numbers are compared, and one of the first and second tables is selected according to a predetermined criterion. The file access system according to claim 2, wherein a position of a directory entry of a file having the first or the second neighboring file number is acquired from the file. The second table is
If there is a table entry for dividing a folder having a file number exceeding a predetermined threshold and there is a folder having a file number exceeding the predetermined threshold, the file belonging to that folder, The file access system according to claim 3, further comprising a table entry of a file whose interval is in a range not exceeding the predetermined threshold. The second table is
The file access system according to claim 3, wherein the file access system has table entries having a predetermined number of file number intervals without having the number of table entries corresponding to the number of all file numbers.

Images