JPH07169196A - Recording and reproducing method in disk device - Google Patents

Abstract

PURPOSE:To shorten waiting time without increasing the revolving speed of a motor and to prevent the power consumption of a device from being increased by dividing circular tracks into circular arc-shaped tracks having plural sectors and writing the same data in corresponding sectors in tracks. CONSTITUTION:When the same data are written in plural positions, for example, in two positions of TBS1, TBS2 in one circular track, data are read-out in a period less than one revolution (less than 180 deg.). In general, the number of divisions of a circular track into circular arc-shaped tracks is defined as (k), data can be read-out in a period less than one-k th of a revolution. Further, since the same data are recorded in plural circular arc-shaped tracks A, B in one circular track respectively, when an error is detected from the data read-out from a certain sector of the track A by an inspection by an error detecting means such as an ECC check, a probability that correct data are obtained by reading out data from a sector of the track B in which the same data are written, is increased, therefore the reliability of the device is increased remarkably.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic computer such as a magnetic disk device, a floppy disk device, or a magneto-optical, write-once type, phase change type optical disk device or the like, which uses a disk-shaped disk as a recording medium. The present invention relates to a data recording / reproducing method in a disk device used as an auxiliary storage device.

[0002]

2. Description of the Related Art In recent years, with the increase in the amount of information in the information-oriented society, it has been required to increase the storage capacity and speed of data access of the above-mentioned disk device, which is an auxiliary storage device for electronic computers. ing.

Generally, the speed of data access in this type of disk device is increased by speeding up the positioning of the head and increasing the rotational speed of the spindle motor to shorten the rotational waiting time.

Regarding the rotational speed of the spindle motor,
3,600 rpm was generally used, but in recent years, this number of revolutions has increased to 4,500 rpm or 5,400 rpm, which is about 1.5 times faster than before, and the waiting time for rotation has also been shortened by about 1.5 times. However, an increase in the rotation speed of the spindle motor leads to an increase in the drive current of the motor, and it is not desirable to increase the rotation speed of the motor for a disk device that requires low current consumption. Therefore, there is a demand for a disk device that can reduce the rotation waiting time without increasing the rotation speed of the spindle motor.

A conventional disk device uses one index signal recorded on the disk as a rotation reference of a spindle motor, and a concentric or spiral track is sectored by a sector pulse signal based on this index signal. Is divided into

Therefore, when the data of the target sector is read out, the sector of interest is detected by counting the sector pulse signal with reference to this index signal, and the data of the sector is read out. .

[0007]

However, in the above-mentioned conventional disk device, if the target sector is just at the read position of the head when the read operation is started in response to the read command, the read operation can be performed immediately. Yes, but if this sector is not in the read position of the head, you must wait until this sector comes to the read position again, and the worst situation is immediately after the target sector has passed the read position, In this case, it is necessary to wait for the data to be read until the target sector reaches the read position again by rotating the disk almost once.
There was a problem that a waiting time close to the amount of rotation was generated.

The present invention has been made to solve the above problems, and an object of the present invention is to set a rotation waiting time until a sector for writing or reading arrives at a position of a head of a spindle motor. It is an object of the present invention to provide a recording / reproducing method for a disk device, which can reduce the number of rotations of the disk drive without increasing and also reduce the frequency of occurrence of read errors.

[0009]

As shown in the principle diagram of FIG. 1, in a recording method in a disk device, a plurality of concentric tracks are divided into arcs of equal length, and the concentric tracks are divided. A disk-shaped disc having a plurality of arc-shaped tracks formed therein is used, and the recording means records the same data on a plurality of arc-shaped tracks belonging to the same circular track of the disk-shaped disc.

The plurality of arc-shaped tracks described above are formatted so as to form a set having the same sector address, and the recording means records the same data in a plurality of sectors having the same sector address. May be.

When this formatting is performed, different identification signals for identifying these arcuate tracks may be recorded on a plurality of arcuate tracks having the same sector address of the disk-shaped disk.

When reproducing from this disc,
When a read error occurs in the data read from a certain sector, the reproducing means can be configured to read again from another sector having the same sector address as this sector.

Instead of the disk-shaped disc D having a plurality of concentric circular tracks, a spiral track having a central angle of 360 / K degrees (K is an integer of 2 or more than 2) and having the same number of sectors is used. A disc-shaped disc having arc-shaped tracks may be used, and in this case as well, the same data may be recorded in each of the plurality of arc-shaped tracks, as described above.

[0014]

As shown in FIG. 1 (a), the disk-shaped disk D is provided with a plurality of concentric circular tracks T1, T2, ..., Each of these concentric circular tracks T1, T2 ,. Are divided into arcs of equal length (that is, the center angles are equal), and a plurality of arc-shaped tracks A and B are formed during formatting.

FIG. 1 (b) shows one circular track of the above-mentioned concentric circular tracks, which is developed linearly. This circular track is an arcuate track A of equal length with the same central angle. It is divided into B, and index pulses INDEX1 and INDEX2 are recorded at the tips of these arc-shaped tracks. Note that in the example shown in the figure, a circular track
The figure shows a case of 180 degrees divided into two, but three divisions, four
It may be divided into more arc-shaped tracks such as division.

The arc-shaped tracks A and B are further divided into a plurality of sectors TAS1, TAS2, ..., TASn, TBS1, TBS2, ... TBSn, and the sector pulse Sct is provided at the head of these sectors.
Is recorded.

When writing data on such a disc-shaped disk, the same data is written on the plurality of arc-shaped tracks A and B. At this time, it is desirable to write the same data to the corresponding sectors of the arc-shaped tracks A and B, for example, TAS2 and TBS2. At this time, the same data is written to the corresponding sectors of these tracks A and B, for example, TAS2 and TBS2. If the sector address of is assigned, the correspondence between TAS2 and TBS2 becomes clear, which is convenient for recording / reproducing processing.

At the time of reproduction, when the read head moves to a position corresponding to a track including a sector for reading data and is in a read state, when the sector to be read comes to this read position, the data is read immediately. However, in general, reading cannot be performed until the sector to be read comes to this reading position, and at the worst timing when the sector to be read enters the reading state immediately after passing the reading position, the disk is not read. Reading cannot be performed until it makes about one rotation.

In the present invention, since the same data is written in a plurality of locations (two locations, TAS2 and TBS2 in the above example) on one circular track, the disk is rotated within one rotation (in this example, in this example). Readings can be taken during rotations (up to 180 degrees). Generally speaking, if the number of divisions of a circular track into an arc-shaped track is K, then 1 / K of the disk
Reading can be performed within the period of rotation.

Further, according to the present invention, as described above,
Since the same data is recorded on each of a plurality of arc-shaped tracks A and B of one circular track, the data read from a sector of one track A is error-corrected by an error detection means such as an ECC check. If detected, the probability that correct data will be obtained by reading the data from the sector in which the same data of the other track B is written becomes extremely high, and therefore the reliability of the disk device is significantly improved.

Generally speaking, this is a circular track
If each arc-shaped track is divided and the number of divisions is K, the probability of an error that occurs during reading is 1 / K as compared with the conventional technique in which the circular track is not divided.

When a disk having spiral tracks is used, the index code is usually recorded on one radial line, so that the track between the index codes (360 ° if the central angle is used). Is regarded as one circular track in the above-mentioned concentric circular track, and this track is divided into equal number of sectors to form an arc-shaped track, the disk having the above-mentioned concentric circular track is used. It will be clear that the same action and effect as the above can be obtained.

[0023]

EXAMPLE FIG. 2 is an expanded view of one circular track for showing the state of the format of a disk having concentric circular tracks used in this example. This one circular track is shown in FIG. Different index signals INDEX1 and INDEX2 are recorded at the head position divided into two by 180 degrees to divide into two arc-shaped tracks, A track and B track, and these two tracks are separated from each other due to the difference in the index signal. Be identifiable.

These tracks A and B are sector pulse signals.
By recording Sct at the beginning of each sector, A
The track is B in n sectors of TAS1, TAS2, ..., TASn.
The tracks are n sectors TBS1, TBS2, with the same number as the A track.
......, TBSn. Then, the sectors corresponding to the A track and the B track, for example, TAS1 and TBS1, TAS2 and TBS
The same sector address is assigned to 2, ...

Each sector is composed of a servo information area, an ID area and a data area which are separated from each other by a gap. In this servo area, a cylinder and a sector number are recorded as information directly related to the present invention. In the ID area, there is flag information in addition to information such as a logical sector address (LSN), a cylinder, a head, etc., in which the head of the logical block of the track is set to “0” and a continuous number is assigned to subsequent logical blocks. As the flag information, a flag indicating a defective sector or a replacement sector and a track identification flag for identifying the A track and the B track according to the present invention are recorded.

As the sector number, the sectors corresponding to the A track and the B track, for example, TA
The same sector number is assigned to S1 and TBS1, TAS2 and TBS2, .... Needless to say, no data is recorded in the data area in the formatted state.

If there is no defective sector in the track having the format shown in FIG. 2, the logical block address (LBA) and the logical sector address (LSN) are assigned to the physical sector as shown in FIG. Note that FIG. 3 shows an example in which there are two heads, the number of cylinders for recording data is three, and the number of sectors in the entire track is 16 sectors. Further, a replacement cylinder including only replacement sectors is further provided. Although not shown in this figure.

The sector indicated by "S" in the column of logical block address (LBA) in FIG. 3 is a substitute sector provided for each cylinder, and a defective sector exists in the sector in this cylinder. Sometimes, this alternative sector is used instead of the defective sector, and when all of these alternative sectors are used and the number of alternative sectors is still insufficient, the sector of the replacement cylinder is used.

However, as shown in FIG. 3, if the sectors in each track and each cylinder are sequentially arranged from the position of the index signal, for example, Japanese Patent Laid-Open No. 2-31 will be used.
As described in Japanese Patent No. 0859, a head skew or a cylinder skew for switching heads or cylinders when reading data which is continuously written in sectors read by different heads or other cylinders. Due to this, a delay time is generated and a waiting time close to one rotation of the disk is generated.

In order to avoid this waiting time, as shown in FIG. 4, the angular position of the sector on another track is shifted by 2 sectors as the track skew period corresponding to the head switching, and the cylinder A cylinder skew period for 3 sectors is provided as a time corresponding to the switching.

FIG. 5 is a diagram showing an alternative process when a defective sector occurs in the cylinder 1 shown in FIG.
As shown in the figure, when it is detected that the logical block addresses 17 and 19 of the A track are defective sectors at the time of formatting, the same logical block address of the B track is detected.
Even if 17 and 19 are not defective sectors, they are managed as defective sectors, and when it is detected that the B track logical block addresses 24 and 25 are defective sectors, the A track logical block addresses 24 and 25 are similarly processed. Manage as a defective sector.

Since the A track and the B track are managed by the same sector address, there is a contradiction if the sector 17 of the A track is replaced but the sector 17 of the B track is not replaced. As such, such management is performed. If a corresponding table for managing where the same data is recorded is provided for each arcuate track, the above management is not necessary.

When a defective sector is generated in this way, sector slip processing is performed, and as shown in (b), the logical sector becomes a normal sector until all the replacement sectors provided in this cylinder are used up. Block addresses are sequentially assigned.

However, when there are four defective sectors, which is larger than the three replacement sectors provided in each cylinder as in this example, as shown in FIG. One defective sector that could not be accommodated is replaced by the sector of the replacement cylinder.

FIG. 6 shows an embodiment of a disk device for writing and reading the disk 1 formatted as described above. The disk 1 is driven by a spindle motor (not shown) at a predetermined speed of 3,600 rpm, for example. The head 2 is provided for writing / reading to / from this disk.

It should be noted that the head 2 is advantageous in that a write head for writing and a write head for reading are separately provided adjacent to each other in order to immediately read the written data and perform verification. For the sake of convenience, the following description will be made on the assumption that the head performs both writing and reading.

The head 2 is connected to a read / write circuit (R / W circuit) 4 via a preamplifier 3, and a disk controller (DC) 5 is a microprocessor unit () operated by a built-in control firmware. Hereinafter, the writing and reading of data to and from the disk 1 is controlled under the control of the MPU 7) in accordance with a command sent from the host 8. The command analysis circuit 9, the R / W control circuit 10, It comprises a buffer RAM 11 for temporarily storing data, and a buffer management control circuit 12 for controlling the buffer RAM 11.

A process of writing data on the disk formatted as described above by the disk device of this embodiment will be described.

Since the host 8 designates the access target of the disk by the logical block address (LBA), the disk controller (DC) 5 receives this logical block address from the host 8 received via the interface (I / F). "Cylinder, head, logical sector address (LSA)
It is converted to an internal logical address "" to access the disk.

This conversion cannot be performed by a simple operation due to defective sectors and skew, and the MPU 7
It is executed using two types of management tables, a defect table and a cylinder skew table, which are managed by the firmware. The defect table manages the location of the defective sector as "head number, sector number" information for each cylinder, and the cylinder skew table determines the cylinder skew determined according to the number of replacement sectors for each cylinder. It manages information about values.

The host 8 issues a command for instructing write or read following the logical block address of the sector to be accessed, and the disk controller 5 analyzes this command by the command analysis circuit 9 therein. Assuming that a write command has arrived here, the command analysis circuit 9 outputs the MPU 7
Is notified of the start of the write operation, and the write data sent from the host 8 following this command is passed from the command analysis circuit 9 through the buffer management control circuit 12 to the buffer RA.
Once stored in M11.

When the MPU 7 receives the write operation start signal from the command analysis circuit 9 described above, it drives a servo mechanism (not shown) to move the head 2 to a cylinder (track) position including a sector to write data. When the positioning of the head 2 is completed, the servo information recorded in each sector of this cylinder is sequentially read by the head 2.

After this read signal is amplified by the preamplifier 3, the servo demodulation circuit 6 demodulates this signal into the sector number and cylinder number of the sector at the read position, and M
PU7 recognizes this sector number and recognizes the index signal IN
Generate DEX and sector signal Sct. Specifically, FIG.
Of the sector SET0-A, the INDEX1 of "0" is generated, and the servo information of the sector SET0-B records the INDEX2 of "1".

Then, the MPU 7 issues a write instruction to the sector to be written to the R / W control circuit 10 inside the disk controller 5, and the R / W control circuit 10 makes an MP.
The sector to be written is detected based on the index signal and sector signal from U7, and the ID is read from the ID area of this sector.
Read the signal.

In the ID area, as described above, the track identification flag for identifying the A track and the B track is recorded according to the present invention. Therefore, if this flag is "0", it is read. If the present sector belongs to track A and this flag is "1", then
R / W control circuit 1 that belongs to a track
0 and MPU7.

The MPU 7 generates the index signal INDEX and the sector pulse signal Sct by recognizing the sector number sent from the servo demodulation circuit 6. For example, an index signal INDEX1 indicating the start position of the track A is generated by the servo information written at the beginning of the first sector TAS1 of the track A, which is the first arc-shaped track, and is the second arc-shaped track. The index information INDEX2 indicating the start position of the track B is generated by the servo information written at the head of the first sector TBS1 of the track B.

The R / W control circuit 10 of the disk controller 5 monitors whether or not it is a sector to write data based on the cylinder number and the sector number.
When receiving a notification from PU7 that it is the target sector, M
The target sector is identified based on the index signal INDEX and sector signal Sct given from PU7, and the I
The R / W circuit 4 is controlled so as to read D (see FIG. 3).

In this ID, as shown in FIG. 3, in addition to the same information as the conventional one such as the cylinder number, the HD (head) number, and the sector number, the presently read track is A, Since a flag indicating which track is B, that is, "0" is recorded for the A track and "1" is recorded for the B track, by reading this flag, the currently passing track is determined. Whether the track is A or B is determined by the R / W control circuit 10 and the MPU.
It can be determined by 7.

When the R / W control circuit 10 detects that the sector at the write position of the head 2 is the sector to be written, the data to be written is read from the buffer RAM 11 via the buffer management control circuit 12.
After converting this data into a recording code by the R / W circuit 4,
The disc 1 is sent to the write head 2 via the preamplifier 3.
Write to.

At this time, the R / W control circuit 10 and the MPU 7 discriminate whether the sector written as described above belongs to the A track or the B track. When the data is first written in the sector to which it belongs, the same data as the data written in the A track is written in the corresponding sector of the B track after the writing in the A track is completed.

On the contrary, when the sector belonging to the B track is written first, the same data as the data written in the B track is also written in the corresponding sector of the A track after the writing in the B track is completed. Write. The second writing process is performed in the same manner as the first writing process.

To give a concrete example, if the target sector to which data is to be written is the second sector of tracks A and B, the head is set to track A including these sectors.
After moving it to B, the desired data is written in the first sector to be written by the rotation of the disk, for example, the second sector TAS2 of the A track, and the second sector TBS2 of the B track is written in the disk. The same data as the sector TAS2 is written in this sector TBS2 after waiting for the head position to be reached by the rotation of.

As described above, in order to write the same data as the first writing to the B or A track after the first writing to the A or B track, the contents in the buffer RAM 11 are written even after the first writing is completed. It is necessary to leave the write data of as it is. It should be noted that it is desirable to perform a verify process of reading the written data with the read head immediately after writing and comparing the written data with the written data. It is desirable to keep the write data.

When the data written as described above is read, since the same data is written in the same logical sector address in track A and track B, it is designated whether track A or track B. It is not necessary to read out by distinguishing or distinguishing, and the data of track A or track B in which the sector to be read is first detected may be read as it is. As a result, the rotation waiting time is reduced to half of the conventional one.

If an error is found in the read data, the data of the corresponding sector (sector having the same sector address) of the read track, for example, the other B track on the opposite side from the A track is read. Read it again. Therefore, even if an error has occurred in one track A or B, the data in the other track B or A can be used, and the error occurrence frequency can be halved as compared with the conventional case. The presence / absence of this error is determined by the error detection code E included in the data.
It can be detected by CC (see FIG. 3).

Although the disk having the concentric circular tracks is used in the above-mentioned embodiment, a disk having a spiral track can be used. In this case, for example, an index code is formed on one radial line. Is recorded, and a track having a central angle of 360 ° between the index codes is treated as one circular track of the concentric circular tracks in the above-described embodiment, and this track is divided into equal sectors to form an arc shape. It will be apparent that the formation of the tracks provides the same action and effect as the case of using the disk having the concentric tracks described above.

Further, in the above-mentioned embodiment, A track and B track
Although the same sector address is given to the sector corresponding to the track, if a correspondence table for managing where the same data is recorded is provided for each arcuate track, A track and B track can be set as described above. It is not necessary to give the same sector address to the sector corresponding to the track.

[0058]

As described above, according to the method of the present invention, the circular track is divided into a plurality of arc-shaped tracks each having an equal number of sectors, and the corresponding arc-shaped tracks in each of the plurality of arc-shaped tracks are divided. Since the same data is written in the sector, the rotation waiting time can be shortened and the error occurrence frequency can be reduced without increasing the rotation speed of the spindle motor. Further, since it is not necessary to increase the number of rotations of the spindle motor, the power consumption of the disk device does not increase and the size can be reduced.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a track format according to an embodiment.

FIG. 3 is a diagram showing a basic sector arrangement.

FIG. 4 is a diagram showing sector arrangement in which skew is taken into consideration.

FIG. 5 is a diagram showing replacement of sectors by replacement processing.

FIG. 6 is a block diagram of an embodiment of a disk device to which the method of the present invention is applied.

[Explanation of symbols]

A first arc-shaped track (A track) B second arc-shaped track (B track) TAS1 to TASn A track sector TBS1 to TBSn B track sector 1 disk 2 head 3 preamplifier 4 read / write circuit (R / W circuit) 5 Disk controller (DC) 6 Servo demodulation circuit 7 Microprocessor unit (MPU) 8 Host 9 Command analysis circuit 10 Read / write control circuit (R / W control circuit) 11 Buffer RAM 12 Buffer management control circuit

Claims (6)

[Claims] 1. A disc-shaped disc in which a concentric circular track is divided into arcs to form a plurality of arc-shaped tracks, recording means for recording data on the disc-shaped disc, and recording is performed on the disc-shaped disc. A recording method in a disk device comprising a reproducing means for reproducing data, wherein the recording means records the same data on a plurality of arc-shaped tracks belonging to the same concentric track. Recording / reproducing method in device. 2. A spiral track having a central angle of 360.
/ K degrees (K is an integer greater than or equal to 2) and a disc-shaped disc in which arc-shaped tracks each having the same number of sectors are formed, recording means for recording data on the disc-shaped disc, and recording on the disc-shaped disc. And a reproducing means for reproducing the recorded data, wherein the recording means records the same data on each of the plurality of arc-shaped tracks. Method. 3. The disk-shaped disk is formatted so that a plurality of arc-shaped tracks form a set having the same sector address, and the recording means writes the same data to a plurality of sectors having the same sector address. The recording / reproducing method in the disk device according to claim 1 or 2, wherein recording is performed. 4. A plurality of arc-shaped tracks having the same sector address of the disk-shaped disk are formatted by recording different identification signals for identifying these arc-shaped tracks. Item 3
A recording / reproducing method in the disk device described. 5. The reproducing means re-reads from another sector having the same sector address as this sector, when a read error occurs in the data read from a certain sector. Recording / reproducing method for disk device. 6. A recording / reproducing method in a disc device comprising a disc-shaped disc on which the same data is recorded on a plurality of arc-shaped tracks, and a reproducing means for reproducing the data recorded on the disc-shaped disc. When the error occurs in the data read from one arc-shaped track, the reproducing means reads the same data as the error-recorded data recorded in another arc-shaped track. Recording / reproducing method for a disc device.