JPS6139286A - Data control method - Google Patents

Abstract

PURPOSE:To enable the titled method to immediately select one of multiple-written plural data blocks by comparing and collating output data with a counter and a register with a collating circuit while control information is written on respective blocks and is read. CONSTITUTION:Control information is recorded to a data block number recording area (c). Namely, the area (c) consists of 3 parts, and first 2 parts are a track number recording part (f) and another is multiple data recording part (g). Control data are stored from reading signal data outputted from EOR9 to a register 10 for reading. At this time, A counter 11 sets preset value from MPU1 beforehand, when reading and reading errors are executed, they jog by the pulse from MPU1, these values are outputted to a comparing circuit 12 each time they jog, and the values are compared and collated with the output data of the register for reading at the comparing circuit 12.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a data management method when the same data is multiplex-written in a plurality of blocks of a data file having sequential block numbers.

Data written to or read from data files (eg, magnetic disk devices, optical disk devices, etc.) used as external storage devices of information processing systems is required to be highly reliable. Therefore, recording media that do not cause data errors have been developed and various correction methods have been put into practical use when data errors are detected.

However, in high-quality data files with a relatively high data error rate, such as in optical disk devices, the same data is multiple-written to ensure highly reliable data. There is a need for the practical implementation of a flexible management method.

(Prior Art) As a prior art, a data management method for an optical disk device will be described as an example with reference to the drawings.

FIG. 3 is a block diagram of a conventional optical disk device, in which (A) shows the write circuit, (B) shows the read circuit, and FIG. 4 shows the structure of the data block.

This example shows a case where a microprocessor (hereinafter referred to as MPU) 1 controls writing/reading of data on an optical disc. Also, optical disk media.

Data is written/read in units of data blocks a as shown in FIG.

Note that in the first 10 areas of data block a, the trunk number and sector number (in this example, logical sectors are referred to as data blocks), which are the physical management numbers of the data file, are recorded in advance at the time of media creation. ing.

Further, logical management numbers are consecutively assigned to the sectors, and one sector is a data block number recording area C in which the logical management numbers are recorded.

It consists of a disk recording area d for recording user data, and an ECC area e for recording data and error correction codes (hereinafter referred to as IECC).

Conventionally, when data is multiple-written in data blocks (sectors) as described above, data is multiple-written in a predetermined number of data blocks. That is, when the old writing register 3 receives an instruction to output management data for multiplex writing from the MPIll via the MPU hash 2, the writing register 3 transfers the latest stored management data to the multiplexer (hereinafter referred to as MP
(referred to as X).

For example, the latest management data stored in the write register 3, that is, the data block number to be recorded in the data block number recording area C, is output to MP[11, and the management data is newly stored in the write register 3. Updates will be made on the day and evening.

It is assumed that failure recovery management data is stored in the write register 3 at the same time.

The liver × 4 sequentially selects the transfer data from the host device (not shown) and the management data from the Z register 3, outputs it to the internal modulation circuit (not shown), and outputs it to the optical disk. It is recorded on the medium in the format shown in FIG.

For reading data written in the manner described above.

The read signal ■ is demodulated by a variable frequency oscillator circuit (hereinafter referred to as VFO) 5, the serial data with one phase corrected is converted into parallel data by a shift register 6, and then sent to a sector buffer 7 (consisting of two sector buffers). Buffer sequentially.

The read signal (1) data taken out from the sector buffer 7 is sent to an exclusive OR circuit (hereinafter referred to as EOR) 9 and corrected if a data error occurs based on the output from an error correction circuit (hereinafter referred to as ECC) 8.

It is output to a transfer buffer (not shown) and sent to a secondary device (not shown).

On the other hand, the management data is stored in the read register 10 (multiple data can be stored at the same time).
' If even one of the multiple-written data block numbers sent to the MPUI has been read, control is performed so as to shift to the reading of the primary data.

In this way, when multiplex writing is performed on a medium using the conventional method, the multiplicity is always set to a predetermined multiplicity.

[Problem that the invention seeks to solve]

In the case of multiple writing using the method described above, in order to perform constant multiple writing for all data, a large area is required to record one piece of data. Moreover, there is a problem in that it is not possible to cope with a method in which the degree of multiple writing is varied depending on the content of data to be recorded, for example.

Furthermore, since the determination when one of the multiple data blocks that have been multiplexed has been successfully read is determined based on what is left behind with MP[1], extra control time is required.

[Failure to solve the problem]

The present invention aims to realize a novel data management method that solves the above-mentioned problems.

The problem is that management information consisting of a flag indicating multiple writing, the number of times the same data has been written multiple times, and a serial number for multiple programs/7rs that have written the same data multiple times is written on each block, and the When reading Brosoino.

storage means for holding management information of a plurality of blocks in which the same data is recorded from the read management information;

a counting means for counting the additional numbers of the plurality of blocks recording the same data; and a collation for holding the expected additional numbers of the plurality of blocks and comparing and collating the output information of the storage means and the output value of the counting means. The problem is solved by a data management method according to the present invention, which includes a means for checking data, and selects one block among multiple write blocks as a data reading block based on the judgment of the collation means.

[Effect]

That is, a management flag area for determining whether multiple writing is performed on the same track or an alternate trunk area as management data for multiple writing, and an area for recording a take indicating the degree of multiplexing. An area for recording the data block number of the multiplex area is provided.

When reading these management data for multiplex writing, 5 a verification circuit that records the multiplicity of the multiplexed area and the data block number in advance, a counter that counts the management data for multiplex writing, and a record-holding of the management data for multiplex writing are required. By comparing and collating the output data of the counter and the register with a matching circuit, it is possible to determine whether or not there are skipped data blocks, that is, whether or not they can be ignored if there are any skipped data blocks. Furthermore, it is possible to determine the number of data blocks that can be skipped, and to immediately select j of the multiple data blocks written in multiplex.

[Example] The gist of the present invention will be specifically explained below with reference to an example shown in FIGS. 1 and 2.

FIG. 1 shows an embodiment of an optical disk device according to the present invention.
(A) is a reading circuit, (B) is a take block read control situation, and FIG. 2 is a configuration example of a data block number recording area according to the present invention. (A) is a data block number recording area; (B) indicate the management data area.

The same symbols indicate the same objects or contents throughout the figures.

In this embodiment, the data block number recording area C has a second
Record the management data shown in the figure. That is.

Area C consists of three parts, the first two parts are the track number recording part, and the other part is the multiplex data recording part g.
Both are composed of 8 bits.

The first two bits h of the multiplex data recording portion g are management flags indicating the presence or absence of multiple writing. For example, "'00" indicates single writing, and "10" indicates multiple writing.

The next 3 bits j indicate the multiplicity in the case of multiple writing, and the 9th order 3 bits) k is the additional number 1 of the multiple writing data block.
For example, "001" and "002" are recorded.

In the case of single writing, if the data block to be recorded is defective and cannot be recorded, it is recorded in the alternate trunk area (not shown) together with the data block number; however, in the case of multiple writing, the same data block It shall be recorded on the track.

Store the above management data in the write register 3, and then write the latest stored data to the MP according to the MIIUI instructions.
Output to X4 and write as management data.

The MPUI takes in this latest management data. At the same time, the old register 3 is updated to the next latest data.

Next, the management data written using the above method is read using the following method. In this embodiment shown in FIG. 1 (1B), the multiplicity is [4
-1 case is shown. Also, (al~(d+) in (engineering) is the same data multiple writing part, 1 to 4 in the [-block serial number] part in (2) is the serial number of the data block where the same data is multiple written, and (3) is the serial number of the data block where the same data is multiple written. In the "read" part, x indicates a part where the data block could not be read, and ○ indicates a part where the data block could be read.

Read output from EOR 9 (No. g ■ Store management data from data in read register 10. At this time, counter 11 presets the precent value from M P I J I, and there is a read and read error. case.

Steps by a pulse from MPUI. These values are
.

Each time it advances, it is output to the conveyor circuit 12, and the conveyor circuit 12 compares it with the output data of the register 10 for picking.

For example, in (a) of FIG. 1B, if normal reading is possible with block serial number 4, reading register 10 outputs this read serial number 4 to conveyor circuit 12.

On the other hand, the counter 11 is set to "1" in advance from the MPIll, is counted to "4" by the read error pulse and the normal read pulse, and this value is similarly set to "1" by the conveyor circuit 1.
Output to 2.

The conveyor circuit 12 is connected to the register 10. By inputting the same "4" from the counter 11, information that the comparison and verification match is outputted to the liver 01. The liver υ1 thereby sets the counter 11 to "1" again.

Note that the conveyor circuit 12 has a multiplicity and each part (al to
fd) management data such as data block numbers are stored in MPUI.
Assume that it is given by

Next, in (b), since the block serial number 2 is successfully read, the output values of the register 10 and the counter 11 match, and the MPUI controls so that reading of the block serial number 3 and subsequent blocks is skipped.

Also, part tc) shows a case where all reading is impossible, and at this time, the count value (expected value) of the counter 11 is "5".
”. On the other hand, the block serial number 1 of the register 10 (d) is assumed to have been read normally and is output to the conveyor circuit 12. The conveyor circuit 12 compares the two pieces of data with the pre-recorded data, determines that one portion has been omitted, and requests the MPUI to reread the portion (C1).

〔Effect of the invention〕

According to the present invention as described above, it is possible to cope with a variable multiple writing method and to perform efficient multiple writing control.

[Brief explanation of drawings]

FIG. 1 shows an embodiment of an optical disk device according to the present invention.
(A) shows the reading circuit, and (B) shows the data block reading control status. FIG. 2 shows a configuration example of a data block number recording area according to the present invention, (A) is a data block number recording area, (
B) is a management data area. Figure 3 is a block diagram of a conventional optical disk device, where (A) is a write circuit and (B) is a read circuit. Figure 4 shows the structure of the data block. are shown respectively. In fig. 1 is MPIJ, 2 is MPII Nomas. 3 is a write register, 4 is MPX. 5 is a VFO, and 6 is a shift register. 7 is a sector haphazard sofa, 8 is an ECC circuit. 9 is EOR, 10 is a read register. 11 is a counter, and 12 is a conveyor circuit. are shown respectively. +2) / 23'4- / 234 723
4-1(a) XX)

Claims (1)

[Claims] In a device that multiplex writes the same data to multiple blocks of a data file with sequential block numbers, a flag indicating multiple writing on each block, the number of multiple writes of the same data, and multiple blocks in which the same data was multiplex written. storage means for holding management information of a plurality of blocks in which the same data is recorded based on the read management information; a counting means for counting the additional numbers of the plurality of blocks, and a collation means for holding the expected additional numbers of the plurality of blocks and comparing and collating the output information of the storage means and the output value of the counting means,
A data management method characterized in that one block among multiple write blocks is selected as a data read block based on the judgment of the collation means.