JP2012009105A - Data tape control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve a problem that it takes a long time to reproduce image data or the like recorded on a data tape for the purpose to back up the data or the like.SOLUTION: A data tape control device records and reproduces data on a data tape 116 by causing a head to move reciprocating between or across the data tracks with respect to a longitudinal direction of the data tape 116 which has plural data tracks for recording data. The data tape control device comprises: a file restoring and rearrangement means 113 that acquires recorded position and recorded direction of a file to be read from a file index 114 acquired from the data tape 116 and determines reading order of the file based on the acquired recorded position and recorded direction; and a file restoring means 119 that reproduces the file to be read in an order rearranged by the file restoring and rearrangement means 113.

Description

  The present invention relates to a data tape control device for recording a data file on a linear access medium such as a data tape, and more particularly, a method for creating an index of a data file and information to be recorded, and a data file is restored using the method. The present invention relates to a data tape control device.

  When reading a data file recorded on a tape medium such as a data tape, the tape device reciprocates along the data track from the front end to the end of the data tape or from the end to the front end of the tape, and straddles the data track. By moving, the head is moved to the start position where the data file is recorded.

  When reading the data file, the tape device reads the data file from the start position of the data file along the recording direction, and moves to the next data track when reaching the leading or trailing end of the data tape, and changes the head moving direction. Invert and continue reading data file.

  When reading a plurality of files from the data tape, the running time of the data tape required to move to the start position of the data file becomes a bottleneck. Various proposals have been made to reduce the time required for reading a plurality of recorded data in a tape device.

  As a technique for shortening the time for reading the data recorded on the data tape, for example, there is a technique disclosed in Patent Document 1. The tape device of Patent Document 1 compares the start position of all data targeted by the next executable command with the current magnetic head position every time the read / write process is completed, and has the shortest distance. The process is optimized by selecting one.

JP 2000-48549 A

  Data recorded on the data track on the data tape is recorded in either the direction from the beginning to the end of the tape or the direction from the end to the beginning of the tape. Therefore, also in reading, the tape device needs to read data by moving the head in the same direction as in recording. Therefore, when calculating the head moving time, it is necessary to consider the moving direction of the head together. However, in the above-described conventional technique, the recording direction is not taken into consideration, and a data tape composed of a plurality of data tracks is used. When reading data, there is a problem that it takes time.

  In order to solve the above-mentioned problems, the data tape control device according to the present invention provides the data tape by performing reciprocal movement of a head in the longitudinal direction of a data tape having a plurality of data tracks for recording data and movement across the data tracks. A data tape controller that records and reproduces data on a data tape, and obtains a file index including recording positions and recording directions of a plurality of files recorded on the data tape from the data tape. And a file restoration order rearranging means for obtaining a recording position and a recording direction of a file to be read from the file index, and determining a file reading order based on the acquired recording position and recording direction, The read in the order sorted by the file restoration sorting means. Comprising a file restoration means for reproducing the target file, the.

  The data tape control device can shorten the reading time by determining the reading order from the recording position and recording direction of the recorded file and reading the data.

The figure which shows the file index item in Embodiment 1. The figure which shows the file index in Embodiment 1. The figure which shows the relationship between the recording position of the file and index in Embodiment 1 The figure which shows the recording file list in Embodiment 1. The figure which shows the structure relevant to the recording of the data tape control apparatus in Embodiment 1. FIG. The figure which shows the structure relevant to the decompression | restoration of the data tape control apparatus in Embodiment 1. The figure which shows the relationship between the recording position of the file in Embodiment 1, and a data track FIG. 6 is a relationship diagram when no reversing operation is performed at the time of forward restoration in the first embodiment. Relationship diagram when performing reverse operation at the time of forward restoration in the first embodiment FIG. 6 is a relationship diagram when the reversing operation is performed at the time of reverse restoration in the first embodiment. Relationship diagram of seek time when restoring without rearrangement in the first embodiment The figure which shows the file list at the time of decompress | restoring without rearrangement in Embodiment 1 Relationship diagram of seek time when rearranging and restoring in the first embodiment The figure which shows the file list | wrist when rearranging and restoring in Embodiment 1 Flowchart of rearrangement algorithm in the first embodiment The figure which shows the restoration file list in Embodiment 1 Explanatory drawing of the first replacement of the rearrangement process in the first embodiment The figure which shows the decompression | restoration list after the first replacement of the rearrangement process in Embodiment 1. Explanatory drawing of the last change of the rearrangement process in Embodiment 1. The figure which shows the decompression | restoration list after the last replacement of the rearrangement process in Embodiment 1.

Hereinafter, embodiments will be described with reference to the drawings.
(Embodiment 1)
FIG. 5 is a block diagram showing a configuration relating to recording in the data tape control apparatus of the present embodiment. FIG. 3 shows the structure of files and indexes when data is recorded on the tape. FIG. 1 shows the structure of a file index created when data is recorded on a tape. FIG. 2 shows a specific example of the file index. FIG. 2 is simplified for the sake of explanation, and only the file path, the start block address, the start direction, the end block address, and the end direction are shown. FIG. 4 shows a recording file list for designating files to be recorded on the tape.

  In FIG. 5, the data tape control apparatus of the present embodiment has a data recording processing unit 108, a data storage medium control unit 102, and a tape drive control unit 104 as a configuration relating to recording.

  The data tape control device of the present embodiment reads the data recorded on the data recording medium 103 based on the recording file list 101, records the read data on the data tape 116, and stores the backup data of the data recording medium 103. create.

  The data recording processing unit 108 records the file specified in the recording file list 101 on the data tape 116 and creates a file index 110. The data recording processing unit 108 includes a recording direction determining unit 117, a file index creating unit 109, a file index 110, and a file writing unit 108.

  The recording position / direction determining means 117 determines whether the running direction of the data tape is the forward direction or the reverse direction when reading the data block from the block address where the data block is recorded. The direction in which the block address value is read from 0 is defined as the forward direction, and the direction read from there to the EOT and reversed is defined as the reverse direction. That is, the forward direction is determined when the quotient obtained by dividing the block address by the number of blocks / track is 0 or an even number, and the reverse direction is determined when the quotient is an odd number.

  File index creation means 109 creates a file index 110 for the data. The file writing means 118 controls the tape drive control unit 104 to write a file on the data tape 116.

  The data storage medium control unit 102 reads and writes the data storage medium 103.

  The tape drive control unit 104 reads / writes the data tape 116. The tape drive control unit 104 includes a seek unit 105 and a tape position acquisition unit 106. The seek unit 105 seeks the data tape 116 to the designated physical block position. The tape position acquisition unit 106 acquires the current physical block position of the data tape 116.

  Hereinafter, an operation at the time of data recording of the data tape control device configured as described above will be described. A process for recording the files shown in the recording file list shown in FIG. 4 on a data tape will be described as an example.

  First, the data recording processing unit 108 seeks the data tape to EOD. Next, the data recording processing unit 108 performs the following processing for the number of files specified in the recording file list 101.

  The file index creation means 109 reads the file from the data storage medium 103 by the data storage medium control unit 102, acquires the file creation date / time, update date / time, access date / time, read-only attribute, size, creator, file name, A new file is registered in the file index 110 together with the file path. Next, the file index creation unit 109 acquires a block address by the position acquisition unit 106 and adds the acquired block address to the file index 110 as a start block address. The file index creating unit 109 determines the running direction of the start block address by the recording position / direction determining unit 117 and adds it to the file index 110 as the start direction.

  Next, the file writing means 118 writes the data file to the data tape 116 in units of data blocks.

  Next, the file index creation unit 109 acquires a block address by the position acquisition unit 106 and adds the acquired block address to the file index 110 as an end block address. The file index creating unit 109 determines the running direction of the end block address by the recording position / direction determining unit 117 and adds it to the file index 110 as the end direction.

  By repeating this operation for the number of files, the data recording processing unit 108 creates the file index 110 at the same time. After all data files are written to the data tape 116, the file index 110 is written to the data tape 116.

  FIG. 7 shows the state of the data tape 116 in which the file is written based on the recording file list 101, and FIG. 2 shows the file index. Next, the relationship between FIG. 2 and FIG. 7 will be described by taking several files as examples. The file with ID 1 has a positive start direction and a positive end direction. The file of ID1 corresponds to the file 207, and indicates a recording state that does not cross the track 201. The ID3 file has a normal start direction and a reverse end direction. The file of ID3 indicates a recording state that extends from the file block 208 in the track 201 to the file block 209 in the track 202 after being folded back by the EOT 206. The ID4 file has a reverse start direction and a positive end direction. The file of ID4 indicates a recording state that extends from the file block 213 in the track 201 to the file block 214 in the track 203 turned back by the BOT 205. The file with ID 5 has a reverse start direction and reverse end direction. The file of ID5 indicates a recording state that does not cross the track 203.

  If the file is written on the data tape as described above, the position where the file is recorded can be known in detail from the file index by using the file path as a key by the start block address, the start direction, the end block address, and the end direction. .

  Next, a method for restoring a file from a data tape will be described. A description will be given by taking as an example processing for extracting the file specified in the restoration file list shown in FIG. 15 from the recorded data tape shown in FIG.

  FIG. 6 is a diagram showing a configuration relating to file restoration processing of the data tape control device of the present embodiment. Components that are the same as those shown in FIG. 5 are given the same reference numerals. The data tape control apparatus according to the present embodiment includes a data restoration processing unit 112, a data storage medium control unit 102, and a tape drive control unit 104 as a configuration relating to file restoration processing.

  The data restoration processing unit 112 collates the files specified in the restoration file list 111 with the file index 114 and determines the recording position and the recording direction, thereby arranging the file restoration order so that the seek distance is minimized. Replace and restore. The data restoration processing unit 112 includes a file restoration order rearranging unit 113, a rearranged restored file list 115, a file index 114, and a file restoring unit 119.

  The file restoration order rearranging unit 113 reads the restoration file list 111, rearranges the order so that the time for seeking the data tape 116 at the time of file restoration is minimized, and creates the rearrangement restoration file list 115. The file index 114 is a file index related to data recorded on the data tape 116, and is obtained from the data tape 116 via the tape drive control unit 104. The file restoring unit 119 restores the data recorded on the data tape 116 by controlling the tape drive control unit 104.

  The specifications of the data tape 116 and the tape drive 107 will be described. Since the unit to seek is the number of blocks, it is expressed as seconds / number of blocks. The distance between BOT and EOT is physically substantially the same for all tracks, but in the actual data tape 116, there is an error between the tracks. However, the error is negligible. For the sake of explanation, the number of blocks between the BOT and EOT of the track is 150,000 blocks, and the seek time is 140 seconds. In this case, 140 ÷ 150,000≈0.00093 seconds / block.

  Subsequently, the inversion operation will be described. Due to the mechanism, the tape drive 107 requires time for stopping and positioning in the reverse direction at the time of reversal. For example, this time is 10 seconds. When the difference between the end block address of the file to be restored first and the start block address of the file to be restored next is the same value with and without inversion, the amount with inversion requires as much as the time required for inversion. It will take extra seconds. Therefore, it is necessary to consider the presence or absence of inversion when determining the restoration order. The number of blocks that can be sought in 10 seconds is 150000/140 × 10 = 10714. In other words, when there is no inversion, if the distance between the end block address of the file to be restored first and the start block address of the file to be restored next is less than 10714, the time can be shortened compared to the combination with the distance of 0 with inversion. become. This value is referred to as an inversion correction distance.

  In addition, when seeking to the block address of another track, the data tape 116 may seek after a little return even if the current position and the seek destination address are in the same direction and do not overlap. is there. That is, inversion may be necessary. This threshold distance is referred to as an inversion threshold distance.

  Here, the seek time when restoring without rearranging will be described. FIG. 11 shows the seek time of the data tape 116 when restoring in the order of the files specified in the restoration file list 101. After restoring the file 401, 79000 blocks seek until the next file block 402, which is the next file, is restored, and the seek time takes about 73 seconds. The next file seeks 44500 blocks by moving from the file block 403 to the file 404, and the seek time takes about 41 seconds. In this way, when restoration is performed without considering the relationship between the position after completion of restoration of the file and the start position of the file to be restored next, the total distance to seek becomes 144000 blocks, and the total time to seek is 133 seconds.

  Next, a case where the files are rearranged and restored will be described.

  The file restoration order rearranging means 113 reads the restoration file list shown in FIG. 16, rearranges the order so that the time for seeking the data tape 116 at the time of file restoration is minimized, and creates the rearrangement restoration file list 115. . FIG. 15 shows a flowchart of the algorithm for rearranging the file restoration order. FIG. 8 shows a seek operation in the case of no inversion at the time of forward restoration. FIG. 9 shows a seek operation in the case of inversion at the time of forward restoration. FIG. 10 shows a seek operation when the end direction of the file to be restored first is different from the start direction of the file to be restored next. FIG. 17 shows combinations and distances from the first file in the restoration file list to the next file to be restored. FIG. 18 shows the rearranged file list after determining the next file to be restored with the shortest seek distance from the first file in the restored file list. FIG. 19 shows the last combination for searching for the shortest distance in the restoration file list and the distance. FIG. 20 shows a rearranged file list after determining the next file to be restored with the shortest seek distance from the first file in the restoration file list. The rearrangement algorithm will be described below with reference to these drawings.

  The data restoration processing unit 112 reads the file index recorded in advance on the data tape 116 into the file index 114, and then reads the restoration file list 111 shown in FIG. 16 into the rearranged restoration file list 115. In addition, the current position of the data tape 116 is assumed to be 0 immediately after rewinding.

  The file restoration order rearranging unit 113 sets −1 as an initial value of the index [i] of the loop 1 and starts processing (S102). Subsequently, the file restoration order rearranging unit 113 determines whether the index [i] is 0 or more (S103). Since the initial value is -1, the file restoration order rearranging unit 113 acquires the current position of the data tape 116 by the tape position acquisition unit 106 and sets 0 to [End block address] (S201). The file restoration order rearranging unit 113 determines the reading direction of the data tape 116 by the recording position / direction determining unit 117 and sets the forward direction to [end direction] (S202). The file restoration order rearranging unit 113 sets 150001 obtained by adding 1 to 150,000, which is the distance between the BOT and the EOT, as the [shortest seek distance] (S105). The file restoration order rearranging unit 113 sets 0, which is the distance from the BOT to the [end block address], as the [start position] (S106). The file restoration order rearranging unit 113 sets the value 0 of [i + 1] to the temporary index [s] (S107).

  Next, the file restoration order sorting unit 113 sets [i + 1] as an initial value to the index [k] as an initial value in loop 2 (S301), and the index [k] of the sorting restoration file list 115 is 0. File [C: \ DATA \ FILE1. [TXT] is searched through the file index 114, and 0 is acquired as [Start Block Address] and the positive direction is acquired as [Start Direction] (S302). The file restoration order rearranging unit 113 sets 0, which is the distance from the BOT to the [start block address], as the [next position] (S303). The file restoration order rearranging unit 113 starts from the current position of the data tape 116, which is the [start position], and [C: \ DATA \ FILE1. 0, which is the distance to [TXT], is set as [seek distance] (S304).

  Subsequently, the file restoration order rearranging unit 113 compares [End direction] and [Start direction], and does not do anything because both are the same in the positive direction (S305). The file restoration order rearranging unit 113 sets the value 0 of the index [k] to the temporary index [s] because 0 of the [seek distance] is less than 150001 of the [shortest seek distance] (S401). 0 of [Seek distance] is set to [Shortest seek distance], and the positive direction of [Start direction] is set to [End direction] (S402).

  Next, the file restoration order rearranging unit 113 returns to loop 2 (S301), increments the index [k], and the index [k] of the rearrangement restoration file list 115 is [C: \ DATA \ FILE3]. . [TXT] is searched from the file index 114, and the reverse direction is acquired with 140000 as the [start block address] and [start direction]. The file restoration order rearranging unit 113 sets 119500, which is the distance from the BOT to the [start block address], as the [next position] (S303). The file restoration order rearranging unit 113 starts from the current position of the data tape 116, which is the [start position], and [C: \ DATA \ FILE3. 119500, which is the distance of [TXT], is set to [Seek distance] (S304).

  Subsequently, the file restoration order rearranging unit 113 compares the [end direction] and the [start direction], and since the direction is different between the forward direction and the reverse direction (S305), the [reverse correction distance] is changed to 119500 of [seek distance]. ] 10714 is added to make 130214 a new [seek distance] (S306). At this time, the file restoration order rearranging means 113 does not do anything because the [seek distance] 130214 is larger than the [shortest seek distance] 0 (S401).

  Again, the file restoration order rearranging unit 113 returns to the loop 2 (S301), and for the files whose index [k] is 2 to 4 in the rearranged restoration file list 115, from the processing (S301) described above (S403). )repeat. As a result of this processing, the temporary index [s] becomes 0, and if the 0th indicated by [i + 1] in the rearranged file list 115 and the 0th indicated by the temporary index [s] are exchanged, the index 0th in the file list 115 is displayed. The file is [C: \ DATA \ FILE1. TXT] is determined (S405).

  Next, the file restoration order rearranging unit 113 returns to loop 1 (S102), and the index [i] is incremented to zero. Since the index [i] is 0 or more (S103), the file restoration order rearranging unit 113 uses the file path [C: \ DATA \ FILE1. TXT] is acquired, and 61000 is acquired as [End Block Address] and the positive direction is acquired as [End Direction] from the file index 114 (S202). The file restoration order rearranging unit 113 sets 150001 obtained by adding 1 to 150,000, which is the distance between the BOT and the EOT, as the [shortest seek distance] (S105). The file restoration order rearranging unit 113 sets 61000, which is the distance from the BOT to the [end block address], as the [start position] (S106). The file restoration order rearranging unit 113 sets the value 1 of [i + 1] to the temporary index [s] (S107).

  Next, the file restoration order sorting unit 113 increments the index [k] in loop 2 (S301), and the file [C: \ DATA \ FILE3] whose index [k] is 1 in the sorting restoration file list 115 is displayed. . [TXT] is searched from the file index 114, and 140000 is acquired as [Start Block Address] and the positive direction is acquired as [Start Direction] (S302). The file restoration order rearranging unit 113 sets 140000, which is the distance from the BOT to the [start block address], as the [next position] (S303). The file restoration order rearranging means 113 starts from [start position] 61000 to [next position] [C: \ DATA \ FILE1. 79000, which is the distance to [TXT], is set as [seek distance] (S304).

  Subsequently, the file restoration order rearranging unit 113 compares [End direction] and [Start direction], and does nothing because it is the same in the positive direction and the positive direction (S305). The file restoration order rearranging unit 113 sets 1 as the value of the index [k] to the temporary index [s] because 79000 of [seek distance] is less than 150001 of [shortest seek distance] (S401). Then, [seek distance] 79000 is set to [shortest seek distance], and the positive direction of [start direction] is set to [end direction] (S402).

  Next, the file restoration order rearranging unit 113 returns to the loop 2 (S301), increments the index [k], and the file [C: \ DATA \ where the index [k] is 2 in the rearrangement restoration file list 115. FILE5. [TXT] is searched from the file index 114, and 375,000 is acquired as [Start Block Address] and the reverse direction is acquired as [Start Direction]. The file restoration order rearranging unit 113 sets 75000, which is the distance from the BOT to the [start block address], as the [next position] (S303). The file restoration order rearranging means 113 starts from [start position] 61000 to [next position] [C: \ DATA \ FILE5. TXT] distance 14000 is set as [seek distance] (S304).

  Subsequently, the file restoration order rearranging unit 113 compares [End direction] and [Start direction], and does nothing because it is the same in the positive direction and the positive direction (S305). Since the [seek distance] 14000 is less than the [shortest seek distance] 79000 (S401), the file restoration order rearranging unit 113 sets the value 1 of the index [k] to the temporary index [s]. [Seek distance] 14000 is set to [Shortest seek distance], and the positive direction of [Start direction] is set to [End direction] (S402).

  The file restoration order rearranging unit 113 returns to loop 2 (S301) again, increments the index [k], and for the files whose index [k] is 3 to 4 in the rearranged restoration file list 115, The processes (S301) to (S403) described are repeated. As a result of this processing, the temporary index [s] becomes 2, and when the first indicated by [i + 1] in the rearranged file list 115 and the second indicated by the temporary index [s] are replaced, the first in the rearranged file list 115 is first. The file is [C: \ DATA \ FILE5. TXT] is determined (S405). The state of this replacement is shown in FIG. 18 as the first replacement 510 of the rearrangement restoration file list 115.

  Similarly, when the loop is continued, the second file in the rearranged file list 115 is [C: \ DATA \ FILE7. TXT], the third file is [C: \ DATA \ FILE3. TXT], the fourth file is [C: \ DATA \ FILE9. TXT]. FIG. 20 shows the rearranged restoration file list 115 after the rearrangement by the file restoration order rearranging unit 113 is completed.

  FIG. 13 shows the seek time of the data tape 116 when restoring in the order of the files specified in the rearrangement restoration file list 115. After the file 401 is restored, 14000 block seeks are required before the restoration of the file block 404, which is the next file, and the seek time takes about 13 seconds. The next file seeks 1500 blocks by moving from the file block 404 to the file 405, and the seek time takes about 2 seconds. When restoration is performed in this way, the total seek distance is 24,000 blocks, and the total seek time is 23 seconds, which can reduce the time by 110 seconds.

  The data tape control device according to the present embodiment can shorten the reading time by determining the reading order from the recording position and recording direction of the recorded file, and reading the data, and backup the video data and the like. It can be applied to a device that performs such operations and is useful.

DESCRIPTION OF SYMBOLS 101 Recording file list 102 Data storage medium control part 103 Data storage medium 104 Tape drive control part 105 Seek means 106 Tape position acquisition means 107 Tape drive 108 Data recording process part 109 File index creation means 110 File index 111 Restoration file list 112 Data restoration Processing unit 113 File restoration order rearranging means 114 File index 115 Rearranged restoration file list 116 Data tape

Claims (3)

A data tape control device for recording and reproducing data to and from the data tape by reciprocating a head in a longitudinal direction of a data tape having a plurality of data tracks for recording data and moving across the data tracks. ,
A file index restoration unit for obtaining a file index including a recording position and a recording direction of a plurality of files recorded on the data tape from the data tape;
A file restoration order rearrangement unit that acquires a recording position and a recording direction of a file to be read from the file index, and determines a reading order of the file based on the acquired recording position and recording direction;
A file restoration unit for playing back the file to be read in the order rearranged by the file restoration rearrangement unit;
A data tape control device comprising: The file restoration order sorting unit
The data tape control device according to claim 1, wherein the file reading order is determined based on a recording position and a recording direction with respect to the file so that the seek time of the file to be read is minimized. A data tape control device for recording and reproducing data to and from the data tape by reciprocating a head in a longitudinal direction of a data tape having a plurality of data tracks for recording data and moving across the data tracks. ,
A file writing unit for recording data on the data tape as a file;
A file index creation unit that creates a file index including a recording position and a recording direction on the data tape when recording the file;
An index writing unit for recording the file index on the data tape;
A data tape control device comprising:

Images