JP2002014774A - Device for recording and reproducing data and method for rerecording processing data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently rerecord data having been recorded. SOLUTION: A control part 22 issues by banks READ command for reading data recorded on a hard disk 30 out to one bank of a buffer memory 15 and a WRITE command for wiring data to the hard disk 30 by reading the data read out to the one bank, so that the data are rerecorded. Here, a command is issued to a different bank after a command to one bank is issued and before the processing of the command is completed. Consequently, when the writing and reading of data to and from one bank are completed, the writing and reading of data to and from a next bank can speedily be performed, so that data can efficiently be rerecorded.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a data recording / reproducing apparatus and a data re-recording processing method. More specifically, a command for reading data recorded on a randomly accessible recording medium into one of a plurality of storage areas, and writing data to the recording medium using the data read into one storage area are performed. By issuing a command for each storage area, re-recording processing using data recorded on a randomly accessible recording medium is performed, and the command is issued after the command is issued to one storage area. By issuing a command to a different storage area before the completion of the process, the re-recording process is performed efficiently.

[0002]

2. Description of the Related Art Conventionally, material data such as a moving image is recorded on a randomly accessible recording medium such as a hard disk, and a desired portion is selected from the material data recorded on the recording medium to create a new data file. Non-linear editing and the like are performed. If such editing of data, creation of files, or erasure of unnecessary files and data is repeated, the unrecorded area on the recording medium is subdivided. Further, as the unrecorded area is further subdivided, data to be newly recorded is distributed and recorded on the recording medium. For this reason, when data is recorded in an undivided unrecorded area or when data recorded in a distributed manner is read, for example, a seek time for moving a signal recording / reproducing head to a desired position on a hard disk, a desired time, etc. , A rotation waiting time occurs until the disk is rotated to the position, and it becomes impossible to read out desired data efficiently.

For this reason, as shown in FIG. 12, for example, a process of reading fragmented data from a recording medium to a buffer memory and relocating the read data to a continuous area, a so-called defragmentation process, is performed. I have.
In this defragmentation process, as shown in FIG. 13, "READ command issuance" for reading data to be defragmented from a hard disk or the like to a buffer memory,
By processing the issued READ command, the RE
"READ processing" for reading the data indicated by the AD command into the buffer memory, and "WRI" for recording the data read into the buffer memory on the hard disk again.
The processing is performed in the order of "TE command issuance" and "WRITE processing" in which the issued WRITE command is executed and the data in the buffer memory is recorded on the hard disk.
The defragmentation of a file for real-time reproduction, such as video data or audio data of a moving image, is also called A / V compilation.

[0004]

By the way, the data amount of a file recorded on a hard disk is often larger than the memory capacity of a buffer memory. This defragmentation (A / V compilation) process cannot be completed only by performing the above operations twice. Therefore, in the defragmentation process, the data re-recording process is repeatedly performed.

[0005] Here, in order to perform defragmentation processing using, for example, one buffer memory, and to efficiently perform re-recording operations that are repeatedly performed, a "WRITE command issuance" is performed without waiting for completion of "READ processing". Is performed, data may be written to the hard disk before the data indicated by the READ command is prepared in the buffer memory, and the data indicated by the READ command may not be recorded correctly. Further, if the “READ command issuance” is performed without waiting for the completion of the “WRITE process”, the data read in the buffer memory may be rewritten to the data indicated by the next READ command.

Therefore, the amount of data used in one relocation is set to be equal to or smaller than the memory capacity of the buffer memory, and "READ command issuance" and "READ processing" are performed as shown in FIG. ”,“ W
After processing is performed in the order of "WRITE command issuance" and "WRITE processing" to complete one re-recording operation, the next re-recording operation is performed.

As described above, when it is assumed that the command issuance and the operation corresponding to the command are repeatedly performed in order,
Since preparation for the next operation is not performed when data is being read or written, the use efficiency of the hard disk is reduced, and the time required for the defragmentation process is increased.

In addition, not only in the defragmentation process, but also in copying recorded data to another position,
Since the process of reading data into the buffer memory and writing the data read into the buffer memory to another area is performed, the processing time becomes longer even in the case of the copy process.

In view of the above, the present invention provides a data recording / reproducing apparatus and a data re-recording processing method capable of efficiently performing data re-recording processing.

[0010]

According to the present invention, there is provided a data recording / reproducing apparatus comprising: a data storage means having a plurality of storage areas for storing data; a data storage means for writing data to a randomly accessible storage medium; Recording medium processing means for reading data stored in the storage medium, a command for reading data stored in the storage medium into one storage area of the data storage means, and a storage medium using the data read to one storage area. A command for writing data is issued for each storage area, and the storage medium processing means and operation control means for controlling the operation of the data storage means are provided. The operation control means includes one storage area of the data storage means. Is issued after the command is issued to the storage area and before the processing of the command is completed. In the data recording / reproducing apparatus, data is written on a randomly accessible recording medium or data recorded on the recording medium is read during the allocated processing period.

Next, a data re-recording method uses a command for reading data recorded on a randomly accessible recording medium into one of a plurality of storage areas and a data read into one storage area. By issuing a command for writing data to a recording medium for each storage area, re-recording processing using data recorded on a randomly accessible recording medium shall be performed. The command is issued to a different storage area after the command is issued to one storage area and before the processing of the command is completed. In the data re-recording method, data is written to a randomly accessible recording medium or data recorded on the recording medium is read during the allocated processing period.

According to the present invention, a READ command for reading data recorded on a randomly accessible recording medium, for example, data recorded on a hard disk, into one back of a buffer memory, and a READ command reading data into one bank. By issuing a WRITE command for writing data to the hard disk for each bank, data copy processing and defragmentation processing are performed. After a command for one bank is issued, before processing of this command is completed, a command for a different bank is issued. A command is issued.

For example, entry data for reading data from a hard disk into one bank and then re-recording the data on the hard disk is sequentially provided for each bank to control the issuance of a command, and the entry information is stored in a FIFO (First Time). (In First Out) data structure, the command issuance state is determined from the oldest entry information, the command is issued for one bank, and before the processing of this command is completed, A command is issued.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a configuration of a system using a data recording / reproducing apparatus using a randomly accessible recording medium, for example, an AV disc recorder for recording and reproducing video and / or audio data using a hard disk.

Video cameras 100 and 110 are connected to the AV disk recorder 10 as data input devices for inputting video and / or audio data and the like. An application device 120 that controls the operation of the AV disk recorder 10 and performs editing processing of recorded video and / or audio data and the like is connected.

FIG. 2 shows the configuration of the AV disk recorder 10. In the input / output processing unit 11a of the AV disk recorder 10, for example, the SMPT
Video and / or audio data (hereinafter referred to as "input data") of material transmitted as a serial digital interface (SDI) format standardized by E259M is subjected to encoding processing and the like to reduce the data amount To be record data WSa. When the reproduction data RSa read from the hard disk 30 is supplied, the reproduction data RSa is subjected to a decoding process, and is output as video and / or audio data. Also, the input / output processing unit 11b performs the encoding process of the input data from the video camera 110 and the decoding process of the read reproduction data RSb in the same manner as the input / output processing unit 11a. The input / output processing units 11a and 11b are provided with control signals CTa and CTb supplied from the control unit 22 described later via the control bus 50.
Is controlled based on the

The recording data WS generated by the input / output processing unit 11a or the input / output processing unit 11b
To one bank of the buffer memory 15. The reproduction data RS read out from the hard disk 30 is stored in the buffer
The data is supplied to the input / output processing unit 11a or the input / output processing unit 11b via the A bus 51.

The DMA bus 51 is connected to the DMA controller 16
And the operation of the DMA controller 16 is controlled based on a control signal CBD supplied from the control unit 22 via the control bus 50.

The recording data WS held in the buffer memory 15 is transmitted to the hard disk 30 via the SCSI bus 52.
Will be recorded. The reproduction data RS read from the hard disk 30 is held in the buffer memory 15 via the SCSI bus 52.

Here, writing of the recording data WS held in the buffer memory 15 to the hard disk 30, reading of data recorded on the hard disk 30, and data transfer between the input / output processing unit 11 and the buffer memory 15 are performed by allocation. The processing is performed in the assigned processing time slot, for example, the assigned time slot of the time slot generated by the time slot generating circuit (not shown).

The SCSI bus 52 is controlled by the SCSI controller 35, and the operation of the SCSI controller 35 is controlled based on a control signal CBS supplied from the control unit 22 via the control bus 50.

An application device 120 is connected to the external interface circuit 21, and receives various control signals CM from the application device 120 and supplies them to the control unit 22 via the control bus 50.

In the control unit 22, the input / output processing unit 11, the DM
The buffer controller 15 controls the buffer memory 15 and the hard disk 30 via the A controller 16 and the SCSI controller 35, compresses the data amount of the input data supplied to the input / output processing unit 11, and records the data on the hard disk 30 as recording data. The data recorded in 30 is decoded and output as video and / or audio data. Furthermore, a defragmentation process is performed to rearrange the recorded data in a continuous area so that the processing on the hard disk 30 can be performed efficiently. In this defragmentation process, a command is supplied from the control unit 22 to the SCSI controller 35 as a control signal CBS.
The defragmentation process is performed by controlling the operation of the buffer memory 15 or the hard disk 30 by using the command from the server 2 as a command corresponding to the buffer memory 15 or the hard disk 30.

Next, a process for defragmenting data recorded on the hard disk 30 will be described. Here, when performing the defragmentation process, assuming that a plurality of banks of the buffer memory 15 are used, data dispersedly recorded on the hard disk 30 is read into one bank, and the read data is read from the hard disk 30. The data is rearranged by writing in a continuous area.

FIG. 3 shows an operation when the defragmentation process is performed by using the two banks A and B of the buffer memory 15 alternately. Here, 1 in the case of performing the defragmentation process
One operation, that is, an operation of issuing a command or an operation of executing an issued command, is performed in one unit time, for example, one time slot.

First, in the time slot TS1, the processing state for the bank A of the buffer memory 15 is set to "READ command issuance (RC-A)". In this “READ command issuance (RC-A)”, a command for reading data to be relocated to be defragmented from the hard disk 30 to the bank A of the buffer memory 15 is transmitted from the control unit 22 via the SCSI controller 35. Issue.

In the time slot TS2, a process corresponding to "READ command issuance" in the time slot TS1 is executed. Here, the processing state for the bank A shifts from “READ command issuance” to “READ completion wait (RD-A)” for confirming the completion of the processing, as shown in FIG.
Note that the processing state for the bank B also transits in the same manner as in FIG.

When the defragmentation process is performed using a plurality of banks, a processing state for a bank used later is determined according to a processing state of a bank used earlier so that a plurality of commands are not issued at the same time. Is set as shown in FIG. That is, when the processing state of the previously used bank X is “READ command issuance (RC-X)”, the processing state of the bank Y used later is set to “R
EAD command issuance (RC-Y) "or" WRITE command issuance (WC-Y) "cannot be used, and" R
If "wait for EAD completion (RD-X)", the processing state for the bank to be used later can be set to the "READ command issue (RC-Y)" state. When the processing state of the previously used bank is “WRITE command issuance (WC-X)”, the processing state for the bank to be used later is changed to “READ command issuance (RC-Y)” or “WRITE command issuance (WC-X)”. WC-Y) and “wait for WRITE completion (WD-X)”, the processing status for the bank to be used later is “READ command issuance (RC-Y)” or “WRITE command issuance ( WC-Y) ”.

For this reason, in the time slot TS2, the processing state for the bank A is "READ completion wait (RD-
A), the processing state for the bank B is set to “READ command issuance (RC-B)”, and a command is issued to read the next data to be relocated from the hard disk 30 to the bank B of the buffer memory 15. .

When reading of data from bank A is completed in this time slot TS2, the processing state for bank A is changed to "WRI" in the next time slot TS3.
Issue TE command (WC-A) ". This "WRIT
In the "E command issuance (WC-A)", a command for recording data read to the bank A in a continuous area of the hard disk 30 is issued from the control unit 22 via the SCSI controller 35. Also, the time slot TS3
Then, the processing according to “READ command issuance (RC-B)” in the time slot TS2 is executed, and the processing state for the bank A is “READ COMMAND WAIT (RD -B) ".

In the time slot TS4, processing corresponding to "WRITE command issuance" in the time slot TS3 is executed. The processing state for bank A is "WR
WR command to confirm the completion of processing from "ITE command issue"
Wait for ITE completion (WD-A) ". Further, when the data reading from the bank B is completed in the time slot TS3, the processing state for the bank A is not a command issuance in the next time slot TS4. WC-B) ”.

In the time slot TS5, a process corresponding to the "WRITE command issuance" in the time slot TS4 is executed. The processing state for bank B is "WR
WR command to confirm the completion of processing from "ITE command issue"
Wait for ITE completion (WD-B) ". Further, in the time slot TS5, the processing state for the bank B is "W
RITE completion wait (WD-B) ", and when the writing of the data read to the bank A is completed in the time slot TS4, the processing state for the bank A is changed to" READ command issuance (RC-A) ”, the next data to be relocated is transferred from the hard disk 30 to the bank A
Issue a command to read the data.

In the time slot TS6, since the processing state for the bank A is "READ completion waiting (RD-A)", when the writing of the data read to the bank B is completed in the time slot TS5, In the next time slot TS6, the processing state for the bank B is set to “READ command issuance (RC-B)”, and a command for reading the next data to be relocated from the hard disk 30 to the bank B in the buffer memory 15 is issued.

As described above, one relocation using the bank A is completed by the processing of the time slots TS1 to T4, and one relocation using the bank B is performed by the processing of the time slots TS2 to T5. Relocation is completed. By repeating the same processing as described below, the data that has been divided and recorded can be rearranged in a continuous area.

It should be noted that "READ completion wait" and "WRITE
If it is not possible to confirm that the reading or writing of data has been completed correctly in the "E completion wait", the command is issued again as shown in FIG. Note that when issuing a command again, the command is issued at a timing that satisfies the condition shown in FIG.

As described above, when a command is executed in one bank, a command is issued to the other bank, and defragmentation is performed using two banks at the same time. Relocation is possible. For this reason, the defragmentation process can be performed more efficiently as compared with a case where one bank is used to perform rearrangement in four time slots or a case where two banks are used alternately to perform rearrangement.

By the way, as described above, if the data read to one bank in one data read is to be recorded again on the hard disk 30, if the amount of data read in one data read is small, During the defragmentation process, the number of times data is written from the buffer memory 15 to the hard disk 30 increases. Therefore, if there is an empty area in the bank when data is read once, the data is read again to increase the amount of data held in the bank and increase the amount of data when writing data. By doing so, the number of recordings may be reduced and the defragmentation process may be performed efficiently.

Next, with reference to FIG. 6, a description will be given of a defragmentation process in the case where data is read twice because the bank has an empty area in one data read. In the time slot TS11, the bank A of the buffer memory 15
The processing status for "Read command issuance (RC-
A) ".

FIG. 7 is a state transition diagram of the processing for the banks A and B. When the "READ command issuance" is performed, if there is a free area in the bank, "the next data is read again". READ command issuance "is performed in the next time slot.

When the process proceeds from the time slot TS11 to the time slot TS12, in the time slot TS12, a process corresponding to the “READ command issuance” in the time slot TS11 is executed, and data to be relocated from the hard disk 30 to the bank A is read. . In addition, since there is an empty area in the bank in one data read, the processing state for the bank A changes to “READ command issuance (RC-A)” again.
Becomes

In the time slot TS13, a process corresponding to the "READ command issuance" in the time slot TS12 is executed. Further, the processing state of the bank A shifts to “READ completion wait (RD-A)” for confirming the completion of the processing according to “READ command issuance” since “READ command issuance” is the second time.

Here, as shown in FIG. 5, the processing state of the previously used bank X is "READ command issuance (R
C-X) ", the processing state for the bank Y to be used later cannot be" READ command issuance (RC-Y) "or" WRITE command issuance (WC-Y) ", and" READ completion wait "is not performed. (RD-X) ", the processing state for the bank Y to be used later can be" READ command issuance (RC-Y) ". For this reason, in the time slot TS13, since the processing state of the bank A is “READ completion waiting (RD-A)”, the processing state for the bank B is set to “READ command issue (RD-B)”.

When data reading from bank A is completed in time slot TS13, the processing state for bank A is changed to "WRITE" in the next time slot TS14.
Command issuance (WC-A). In the time slot TS14, a process corresponding to the "READ command issuance (RC-B)" in the time slot TS13 is executed.

In the time slot TS15, a process corresponding to "WRITE command issuance (WC-A)" in the time slot TS14 is executed. In addition, the processing state for the bank A shifts from “WRITE command issuance” to “WRITE completion wait (WD-A)” for confirming the completion of the processing. Further, in the time slot TS15, since the processing state for the bank A is "wait for WRITE completion (WD-A)", the processing state for the bank B is set to "READ command issue (RD-B)" for the second time.

In the time slot TS16, the processing corresponding to the "READ command issuance (RC-B)" in the time slot TS15 is executed, and the processing state for the bank B is confirmed from the "READ command issuance". The processing shifts to "READ completion wait (RD-B)". further,
In the time slot TS16, the processing state for the bank B is “READ completion wait (RD-B)”. Therefore, when the writing of the data read to the bank A is completed in the time slot TS15, In TS16, the processing state for the bank A is set to "READ command issuance (RC-A)", and a command for reading the next data to be relocated from the hard disk 30 to the bank A of the buffer memory 15 is issued.

When the reading of data from bank B is completed in this time slot TS16, the processing state for bank B is changed to "WRI" in the next time slot TS17.
Issue TE command (WC-B) ". In the time slot TS17, “REA” in the time slot TS16 is used.
D command issuance (RC-B) ".

In the time slot TS18, a process corresponding to "WRITE command issuance (WC-B)" in the time slot TS17 is executed. In addition, the processing state for bank B shifts from “WRITE command issuance” to “WRITE completion wait (WD-B)” for confirming the completion of the processing. Further, in the time slot TS18, since the processing state for the bank B is "WRITE completion wait (WD-B)", the processing state for the bank A is set to "READ command issuance (RD-A)" for the second time.

In the time slot TS19, the processing corresponding to the "READ command issuance (RC-A)" in the time slot TS18 is executed, and the processing state for the bank A is confirmed from "READ command issuance". The flow shifts to "READ completion wait (RD-A)". further,
In the time slot TS19, the processing state for the bank A is "READ completion waiting (RD-A)". Therefore, when the writing of the data read to the bank B is completed in the time slot TS18, At TS19, the processing state for bank B is set to "READ command issuance (RC-A)".

In this way, the time slots TS11 to TS11
One reallocation using the bank A is completed by the processing of T15, and one reallocation using the bank B is completed by the processing of the time slots TS13 to T18.

By performing the same processing hereafter, the "READ command issuance" is performed twice using the two banks, thereby effectively utilizing the memory capacity of the banks and reallocation of one cycle in nine time slots. It can be carried out.

Further, in the above case, the defragmentation process is performed using two banks, but three buffers may be used. Here, FIG. 8 shows an operation when three banks are used. The state transitions of the three banks A, B, and C are the same as those in FIG.
The figure shows the operation in the case where the processing according to the "EAD command issuance" or "WRITE command" requires two time slots. In the time slot TS31, the buffer memory 15
Of the bank A is "READ command issued (RC-A)".

In the time slot TS32, a process corresponding to the "READ command issuance (RC-A)" in the time slot TS31 is executed. Further, based on the determination condition of “command issuance” shown in FIG. 5, the processing state of the bank A used first is “READ command issuance (RC-A)” or “W
RITE command issuance (WC-A) "
In order to read the next data to the bank B to be used,
The processing state for the bank B is set to “READ command issuance (RC-B)”.

In the time slot TS33, since the process corresponding to the “READ command issuance” requires two time slots, the processing state for the bank A is set to “READ completion wait (RD-A)”. For bank B,
"READ command issuance (R
CB) ", and the processing state of the banks A and B is" WRITE command issuance (WC
-A, WC-B) ". Therefore, based on the determination condition of “command issue” shown in FIG. 5, the next data is read out to the third bank C to be used, and the processing state for the bank C is changed to “READ command issue (RC-C)”.
And

When reading of data from bank A is completed in this time slot TS33, the processing state for bank A is changed to "WRI" in the next time slot TS34.
Issue TE command (WC-A) ". Also, "REA
Since processing in response to "D command issuance (RC-B)" also requires two time slots, the processing state for bank B in time slot TS34 is "READ completion wait (RD-B)".
B) ", and the bank C becomes the time slot TS33.
Is being executed in response to the "READ command issuance (RC-C)".

When the data reading from the bank B is completed in the time slot TS34 and the time slot TS35 is reached, the time slot T35 in the previously used bank A is used.
Processing corresponding to "WRITE command issuance (WC-A)" in S34 is being performed, and the processing state of bank A is "READ".
Since the command is not “command issue (RC-A)” or “WRITE command issue (WC-A)”, the processing state of bank B is set to “WRITE command issue (WC-B)”. Also,
At the time slot TS35, the processing state for the bank C is "READ completion wait (RD-C)".

In the time slot TS36, "WRITE
Since the processing in response to the "command issue (WC-A)" requires two time slots, the processing state for the bank A is set to "WRITE completion wait (WD-A)". For bank B, “WRIT” in time slot TS35
E command issuance (WC-B) "is being executed, and the processing states of banks A and B are those of" command issuance (RC-A, WC-A, RC-B, WC-B) ". No state. Therefore, reading of data from bank C is completed in time slot TS35 and time slot TS35
At 36, the processing status for bank C is "WRITE
Command issuance (WC-C) ".

In the time slot TS37, the processing state for the bank B is set to "WRITE completion wait (WD-B)". For bank C, time slot TS36
Is being executed in response to "WRITE command issuance (WC-C)", and the processing status of banks B and C is "command issuance (RC-B, WC-B, RC-C, WC-C).
C) ". For this reason, when the writing of the data read to the bank A is completed in the time slot TS36, in the next time slot TS37,
Assuming that the processing state for the bank A is “READ command issue (RC-A)”, a command for reading the next data to be relocated from the hard disk 30 to the bank A of the buffer memory 15 is issued.

By performing the same processing as described above, READ
When a response to a command or a WRITE command requires two time slots, a READ command or a WRITE is sequentially used for each time slot by using three banks.
The E command can be issued, and the defragmentation process can be performed more efficiently by using three banks simultaneously.

Even if the state transition diagram of the three banks A, B and C is as shown in FIG. 7, the defragmentation process can be performed as shown in FIG.

In the time slot TS51, the processing state for the bank A in the buffer memory 15 is set to "READ command issuance (RC-A)". When proceeding from time slot TS51 to time slot TS52, time slot TS52
Then, a process corresponding to "READ command issuance" in the time slot TS51 is executed to read data to be relocated from the hard disk 30 to the bank A. Further, since there is an empty area in the bank in one data read, the processing state for the bank A is "READ command issuance (RC-A)" again.

In the time slot TS53, processing corresponding to "READ command issuance" in the time slot TS52 is executed. Further, the processing state of the bank A shifts to “READ completion wait (RD-A)” for confirming the completion of the processing according to “READ command issuance” since “READ command issuance” is the second time.

Here, the processing state of the previously used bank A is "READ command issuance (RC-A)" or "WRIT
Since it is not "E command issuance (WC-A)", the processing state for the next bank B to be used is set to "READ command issuance (RC-B)".

When data reading from bank A is completed in time slot TS53, the processing state for bank A is changed to "WRITE" in the next time slot TS54.
Command issuance (WC-A). In the time slot TS54, a process corresponding to the “READ command issuance (RC-B)” in the time slot TS53 is executed.

In the time slot TS55, a process corresponding to "WRITE command issuance (WC-A)" in the time slot TS54 is executed. In addition, the processing state for the bank A shifts from “WRITE command issuance” to “WRITE completion wait (WD-A)” for confirming the completion of the processing. Further, in the time slot TS55, since the processing state for the bank A is "WRITE completion waiting (WD-A)", the processing state for the bank B is set to "READ command issue (RD-B)" for the second time.

In the time slot TS56, the processing corresponding to the "READ command issuance (RC-B)" in the time slot TS55 is executed, and the processing status for the bank B is confirmed from "READ command issuance". The processing shifts to "READ completion wait (RD-B)". further,
In the time slot TS56, since the processing state for the bank B is “READ completion wait (RD-B)”, when the writing of the data read to the bank A is completed in the time slot TS55, In TS56, the processing state for the bank C is set to “READ command issuance (RC-C)”, and a command for reading the next data to be relocated from the hard disk 30 to the bank C of the buffer memory 15 is issued.

When the data reading from the bank B is completed in the time slot TS56, the processing state for the bank B is changed to “WRI” in the next time slot TS57.
Issue TE command (WC-B) ". In the time slot TS57, “REA” in the time slot TS56 is used.
D command issuance (RC-B) ".

In the time slot TS58, a process corresponding to "WRITE command issuance (WC-B)" in the time slot TS57 is executed. In addition, the processing state for bank B shifts from “WRITE command issuance” to “WRITE completion wait (WD-B)” for confirming the completion of the processing. Furthermore, in the time slot TS58, the processing state for the bank B is "WRITE completion wait (WD-B)", so that the processing state for the bank C is the second "READ command issuance (RD-A)".

In the time slot TS59, the processing corresponding to the "READ command issuance (RC-C)" in the time slot TS58 is executed, and the processing status for the bank C is confirmed from "READ command issuance". The flow shifts to "READ completion wait (RD-A)". further,
In the time slot TS59, the processing state for the bank C is "READ completion wait (RD-A)". Therefore, when the writing of the data read to the bank B is completed in the time slot TS58, this time slot In TS59, the processing state for the bank C is set to "READ command issuance (RC-A)". Hereinafter, by performing the same processing, the defragmentation processing can be performed using the three banks.

When the state of each bank is controlled using, for example, three banks as described above, entry information indicating which bank is in which processing state is stored in the control unit 22 in a FIFO (First Time). By managing the data structure as (In First Out), the operation shown in FIGS. 8 and 9 can be easily performed.

FIG. 10 is a flowchart showing the operation of managing the entry information. When a fragmentation area is generated in the hard disk 30 and the defragmentation process for this area is started, in step ST1, the entry information in the FIFO is changed based on the state transition diagram of FIG. Then, the process proceeds to step ST2.

In step ST 2, the state of “waiting for WRITE completion” is changed to “END” in step ST 1 to detect an area of the entry information for which processing has been completed, and this area in the FIFO is released. ST
Proceed to 3.

At step ST3, it is determined whether or not data to be defragmented remains. If data remains, the process proceeds to step ST4, and if not, the process proceeds to step ST6.

At step ST4, the area of the entry information is changed to the FIFO to perform the defragmentation processing of the remaining data.
It is determined whether or not O can be secured. If it can be secured, the process proceeds to step ST5, and if it cannot be secured, the process proceeds to step ST6.

At step ST5, an area for entry information is secured in the FIFO, and the entry information is allocated to this area. Here, when the defragmentation process is started, entry information is first allocated to the FIFO in step ST5, and the process proceeds to step ST6. Here, since the defragmentation process is performed using the three banks of the buffer memory 15, the FIFO is configured so as to secure three entry information areas so that the processing state of each bank can be managed. When the bank is used in the order of the banks A, B, and C in the buffer memory 15, the entry information indicating that the processing state for the bank A is "READ command issued" is FIFO.
Is allocated to the first area.

At step ST6, it is determined whether or not an area for entry information is secured in the FIFO. If it is determined that an area for entry information is secured, the process proceeds to step ST7.

In step ST7, the entry information of the oldest area in the FIFO is set as entry information for command issue determination, and the flow advances to step ST8.

In step ST8, it is determined whether or not the entry information for command issuance determination is "command issuance". Here, when it is determined that the state is the command issuing state, the process proceeds to step ST9, issues the command indicated by the entry information, and returns to step ST1. That is, when the entry information indicates, for example, that the processing state for the bank A is “READ command issued”, a READ command is issued and the process returns to step ST1. As described above, the time slot TS shown in FIG.
The processing of 51 is performed.

FIG. 11 shows the entry information in the control section 22 at the timing of step ST1. In the time slot TS31, only the entry information for the bank A is allocated to the FIFO. In FIG. 11, “READ command issuance (R
C) and WRITE command issuance (WC)
This indicates that it is assigned to the FO and is not being executed, indicating that “READ command issuance (RC)” and “WRITE command issuance (WC)” surrounded by broken lines have been executed.

When "READ command issuance (RC-A)" is executed for bank A and the process returns to step ST1, in the next step ST1, the entry information is stored as described above.
The transition is made according to the command issuing status or the data transfer status. Here, since data is read twice,
The entry information indicates the processing state for bank A as “READ”.
Command issuance (RC-A) "
The process proceeds to T2 and proceeds to step ST3 because there is no entry information area for releasing the area.

When data to be defragmented remains, the process proceeds from step ST3 to step ST4. Since there is only one entry information area, it is determined that a new entry information area can be secured, and Proceed to ST5.

In step ST5, an area for entry information is newly secured in the FIFO, and the entry information is allocated to this area. That is, the entry information indicating that the processing state of the bank B used next to the bank A is “issue READ command” is allocated to the newly secured area.

Thereafter, from step ST6 to step ST
In step ST7, the entry information of the oldest area in the FIFO is set as entry information for command issue determination, and the process proceeds to step ST8.

In step ST8, it is determined whether or not the entry information for command issuance determination is in a command issuance state. Here, the entry information for command issuance determination indicates that the processing state of the bank A is “READ command issuance”, so the process proceeds to step ST9, and the command indicated by the entry information of the oldest area is entered. And returns to step ST1.

As described above, when the processing from step ST1 to step ST9 is performed for the second time, the processing of the time slot TS52 shown in FIG. 9 is performed. Also,
When the processing of the time slot TS52 is performed, FIG.
As shown in FIG. 1, the entry information of the banks A and B is allocated to the FIFO.

When "READ command issuance (RC-A)" is executed for bank A and the process returns to step ST1, in the next step ST1, the entry information of bank A is changed to "RE".
AD command issuance "is shifted to the next state," READ completion wait ", and the entry information of bank B, that is," READ command issuance "is not executed. The process proceeds to step ST3 since there is no entry information area for releasing the area yet.

If there remains data for which the defragmentation process has not been completed, the process proceeds from step ST3 to step S3.
The process proceeds to T4, and since there are two entry information regions, it is determined that a new entry information region can be secured, and the process proceeds to step ST5.

In step ST5, an area for entry information is newly secured in the FIFO, and the entry information is allocated to this area. That is, the entry information indicating that the processing state of the bank C used next to the bank B is "READ command issued" is allocated to the newly secured area.

Thereafter, from step ST6 to step ST
In step ST7, the entry information of the oldest area in the FIFO is set as entry information for command issue determination, and the process proceeds to step ST8.

In step ST8, it is determined whether or not the entry information for command issuance determination is in a command issuance state. Here, in the entry information for command issuance discrimination, the processing state of the bank A is "READ completion wait (RD-
A) ”indicates that step ST
Go to 10.

In step ST10, it is determined whether or not the area of the next oldest entry information is reserved in the FIFO. If not, the process returns to step ST1.
If it is secured, the process proceeds to step ST11. Here, since the area of the entry information for the bank B is secured after the entry information for the bank A, the process proceeds to step ST11.

In step ST11, the entry information of the next oldest area is set as entry information for command issuance discrimination, and the process returns to step ST8. That is, entry information indicating that the processing state for bank B is "READ command issuance" is set as entry information for command issuance determination, and the process returns to step ST8. For this reason, in the next step ST8, it is determined that the command is in the issued state, and the process proceeds to step ST9, in which the command indicated by the entry information, that is, “READ command issuance (RC-A)” for the bank B is executed, and the process proceeds to step ST1. Return. By the above processing, the processing of the time slot TS53 shown in FIG. 9 is performed. When the processing of the time slot TS53 is performed, the entry information of the banks A, B, and C is allocated to the FIFO, and 3
When one entry information is allocated to the FIFO, the process proceeds from step ST4 to step ST6.

When the processing of the time slot TS55 is performed, the processing state for the bank A changes from “waiting for WRITE completion (WD-A)” to “END”. Is released, and the area of entry information for bank B becomes the oldest. Further, a new entry information area for the bank A is secured after the entry information area for the bank C.

Thereafter, when there is no more data to be defragmented, the process proceeds from step ST3 to step ST6,
When the processing state of each bank becomes “END” and the entry information area is released, and it is determined in step ST6 that the entry information area is not secured in the FIFO, the defragmentation processing is completed.

As described above, the entry information is managed by the data structure of the FIFO, the contents of the entry information are discriminated from the oldest one, the control is performed so that the command is not issued simultaneously, and the processing state of the bank is changed. If the entry information area indicating completion is sequentially opened and used as a new entry information area, the defragmentation process using three banks shown in FIG. 9 can be easily executed. It should be noted that the processing in step ST1 in FIG. 10 is changed to change the execution of “READ command issuance” shown in FIG. 11 from two times to one time and “command completion wait” is set to two time slots. The defragmentation process shown in FIG. 8 can also be easily performed.

If the entry information stored in the control unit 22 is stored so as not to be erased even if the supply of power to the AV disk recorder 10 is stopped during the defragmentation process, the supply of power is stopped. Thus, even if the defragmentation process is interrupted, the interrupted defragmentation process can be resumed using the stored entry information.

In the case described above, the AV disk recorder 10 has a plurality of input / output processing units 11 and the hard disk 3
A time slot is assigned to each input / output processing unit 11 so that data transfer between the corresponding input / output processing unit 11 and the hard disk 30 can be performed in the assigned time slot so that data transfer between the input / output processing units 11 can be performed correctly. In order to perform the defragmentation process, the processing state of the bank is changed every time slot. However, the entry information is managed by the data structure of the FIFO, and the contents of the entry information are discriminated from the oldest one so that the commands are not issued at the same time. After the command is issued and before the processing for the command is completed, the command for the next bank to be used is issued, and the area of the entry information for which the processing for one bank is completed is sequentially opened to newly store the area for the new entry information. As a result, even in a data processing device that does not use a time slot, the defragmentation process can be efficiently performed in the same manner as described above.

For example, in a defragmentation process in a computer device or the like, if relocation of data is performed using a plurality of banks and control is made so that commands are not issued at the same time using entry information, during command processing, Since the next command can be issued, defragmentation can be performed efficiently.

It is needless to say that a plurality of buffer memories may be provided in place of the plurality of banks of the buffer memory 15 and the defragmentation process may be performed by using the buffer memory in the same manner as in the above-described bank.

In addition to the defragmentation process of reading out the subdivided data and rearranging it in a continuous area,
In the case of copy processing for recording data recorded on the hard disk 30 again in another area, the copy processing is efficiently performed by using a buffer memory and a bank and issuing a command in the same manner as in the above-described defragmentation processing. It can be carried out. Further, the data to be re-recorded after the defragmentation process or the copy process is performed is not limited to the data read from the hard disk 30, and is processed or corrected after the read data is processed.
It may be written to 0. Further, the recording medium that can be randomly accessed is not limited to the hard disk 30, but may be an optical disk, a semiconductor memory, or the like.

[0100]

According to the present invention, a command for reading data recorded on a randomly accessible recording medium into one of a plurality of storage areas and recording using the data read into one storage area are provided. A command for writing data to the medium is issued for each storage area, a re-recording process is performed using data recorded on a randomly accessible recording medium, and the command is issued for 1
After the command is issued to one storage area and before the processing of the command is completed, a command is issued to a different storage area. For this reason, when the writing or reading of data to one storage area is completed, the writing or reading of data to the next storage area can be performed promptly, and the re-recording process can be performed efficiently.

Also, entry information for reading data from the recording medium to one storage area and then re-recording the data on the recording medium is sequentially provided for each storage area, command issuance is controlled, and this entry information is stored in a FIFO. (First I
n First Out), the command issuance state is determined from the oldest entry information, and after the command is issued to one storage area and before the processing of the command is completed, a different storage area is used. Is issued. Therefore, it is possible to easily control the issuance of the command.

Further, after the command is issued and the data recorded on the recording medium is read out to the storage area, it is re-recorded at a position on the recording medium different from that of the read data, so that it is possible to improve the efficiency. Copy processing can be performed. Furthermore, a command is issued, and the data divided and recorded on the recording medium is read out to the storage area and then re-recorded at a continuous position on the recording medium, so that the defragmentation process can be performed efficiently. it can.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a system.

FIG. 2 is a diagram showing a configuration of an AV disk recorder.

FIG. 3 is a diagram illustrating a defragmentation process using two banks.

FIG. 4 is a state transition diagram of a bank.

FIG. 5 is a diagram showing command issuing conditions.

FIG. 6 is a diagram showing a defragmentation process when data is read twice using two banks.

FIG. 7 is a state transition diagram including determination of a free area.

FIG. 8 is a diagram showing a defragmentation process using three banks.

FIG. 9 is a diagram showing a defragmentation process when data reading is performed twice using three banks.

FIG. 10 is a flowchart illustrating an operation of managing entry information.

FIG. 11 is a diagram showing entry information in a control unit.

FIG. 12 is a diagram for explaining a defragmentation process.

FIG. 13 is a diagram showing a processing order in a defragmentation process.

FIG. 14 illustrates a conventional defragmentation operation.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 ... AV disk recorder, 11 ... I / O processing part, 15 ... Buffer memory, 16 ... DMA controller, 21 ... External interface circuit, 22
... Control unit, 30 ... Hard disk, 35 ...
SCSI controller, 50 ... control bus, 51 ...
DMA bus, 52 ... SCSI bus, 100, 11
0: Video camera, 120: Application device

Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat II (reference) G11B 27/034 H04N 5/781 510E 27/036 510F H04N 5/765 5/91 L 5/781 N 5/91 G11B 27 / 02 K 27/08 F term (reference) 5B065 BA01 CA15 CE16 CH05 5C053 FA14 FA23 FA27 HA33 JA01 KA04 KA19 KA24 LA01 LA06 5D044 AB05 AB07 BC01 CC04 DE03 DE12 DE25 DE92 DE96 GK12 5D110 AA13 AA27 AA29 BB20 CA02 CA05 CC06 CK26

Claims (16)

[Claims] 1. A data storage means having a plurality of storage areas for storing data; a recording medium processing means for writing data on a randomly accessible recording medium or reading data recorded on the recording medium; A command to read data recorded on a medium into one storage area of the data storage unit, and a command to write data to the recording medium using the data read into the one storage area, The storage medium processing means and operation control means for controlling operations of the data storage means, wherein the operation control means issues the command to one storage area of the data storage means. Issuing the command to a different storage area after the issuance and before the completion of the processing of the command. Recording and reproducing apparatus. 2. The operation control means according to claim 1, wherein entry information for reading data from said recording medium to one storage area of said data storage means and then re-recording said data on said recording medium is sequentially provided for each of said storage areas. Command issuance, and the entry information is stored in a FIFO (First I
n First Out) data structure, and determine the command issue state from the oldest order of the entry information,
2. The data recording / reproducing apparatus according to claim 1, wherein the command is issued to a different storage area after the command is issued to one storage area of the data storage unit and before the processing of the command is completed. 3. The operation control means issues the command to read data recorded on the recording medium into the storage area, and then reads a position on the recording medium different from the read data. 2. The data recording / reproducing apparatus according to claim 1, wherein the data is recorded again. 4. The operation control means issues the command, reads out the data divided and recorded on the recording medium into the storage area, and re-records the data at a continuous position on the recording medium. 2. The data recording / reproducing apparatus according to claim 1, wherein: 5. A data storage means having a plurality of storage areas for storing data, and recording for writing data on a randomly accessible recording medium or reading data recorded on the recording medium in an allocated processing period. Medium processing means; a command for reading data recorded on the recording medium into one storage area of the data storage means; and a command for reading data to the recording medium using the data read into the one storage area. The storage medium processing unit and an operation control unit for controlling the operation of the data storage unit by issuing a command for performing writing for each of the storage areas, wherein the operation control unit includes one of the data storage units. Issue the command to a different storage area after issuing the command to one storage area and before completing the processing of the command A data recording / reproducing apparatus characterized in that: 6. The operation control means, wherein entry information for reading data from the recording medium to one storage area of the data storage means and re-recording the data on the recording medium is sequentially provided for each of the storage areas. Command issuance, and the entry information is stored in a FIFO (First I
n First Out) data structure, and determine the command issue state from the oldest order of the entry information,
6. The data recording / reproducing apparatus according to claim 5, wherein the command is issued to a different storage area after the command is issued to one storage area of the data storage unit and before the processing of the command is completed. 7. The operation control unit issues the command to read data recorded on the recording medium into the storage area, and then reads a position on the recording medium different from the read data. 6. The data recording / reproducing apparatus according to claim 5, wherein the data is re-recorded. 8. The operation control means issues the command, reads out the data divided and recorded on the recording medium to the storage area, and then re-records the data at a continuous position on the recording medium. The data recording / reproducing apparatus according to claim 5, wherein: 9. A command for reading data recorded on a randomly accessible recording medium into one of a plurality of storage areas, and data on the recording medium using the data read into the one storage area. By issuing a command to write the data for each of the storage areas, re-recording processing using data recorded on the recording medium that can be randomly accessed is performed. A method for processing data re-recording, wherein the command is issued to a different storage area after the command is issued to a storage area and before the processing of the command is completed. 10. Entry information for reading data from the recording medium to the one storage area and then re-recording the data on the recording medium is sequentially provided for each of the storage areas to control issuance of the command, and Information FI
It is managed as a FO (First In First Out) data structure, and the issue status of the command is determined from the oldest order of the entry information, and after the command is issued to the one storage area and before the processing of the command is completed. 10. The data re-recording processing method according to claim 9, wherein the command is issued to a different storage area. 11. A method of issuing the command to read data recorded on the recording medium into the storage area, and then re-recording the data at a position on the recording medium different from the read data. 10. The data re-recording processing method according to claim 9, wherein: 12. The method according to claim 12, wherein the command is issued to read out the data divided and recorded on the recording medium into the storage area, and then re-record the data at successive positions on the recording medium. Item 10. A data re-recording processing method according to Item 9. 13. Writing data to a randomly accessible recording medium or reading data recorded on the recording medium during the allocated processing period, and recording the data on the randomly accessible recording medium. A command to read data into one of a plurality of storage areas, and a command to write data to the recording medium using the data read into the one storage area,
By issuing the command for each storage area, a re-recording process using data recorded on the recording medium that can be randomly accessed is performed. A data re-recording processing method, wherein the command is issued to a different storage area later before the processing of the command is completed. 14. Entry information for reading data from the recording medium to the one storage area and then re-recording the data on the recording medium is sequentially provided for each of the storage areas to control issuance of the command, and Information FI
It is managed as a FO (First In First Out) data structure, and the issue status of the command is determined from the oldest order of the entry information, and after the command is issued to the one storage area and before the processing of the command is completed. 14. The data re-recording processing method according to claim 13, wherein the command is issued to a different storage area. 15. A method of issuing the command to read data recorded on the recording medium into the storage area, and then re-recording the data at a position on the recording medium different from the read data. 14. The data re-recording processing method according to claim 13, wherein: 16. The method according to claim 15, wherein the command is issued to read out the data divided and recorded on the recording medium into the storage area, and then re-record the data at a continuous position on the recording medium. Item 14. A data re-recording processing method according to Item 13.