KR100464410B1 - Reassignning method of defect sector and access method for defect sector therefor - Google Patents

Abstract

The present invention relates to a method for reallocating a defect sector and a method for accessing a defect sector suitable for the method.

A defect sector reallocation method according to the present invention is a method for processing a defect sector in a headerless ID hard disk drive. Reading out to the end and storing it in a buffer; Storing the contents stored in the buffer sequentially from the next sector of the defective sector to the spare area; And creating a reassignment list having sector pulse information from which sector pulses of the defect sector have been removed.

According to the defect sector reallocation method according to the present invention, there is an effect of minimizing the execution speed reduction by the reallocation sector.

Description

Reassignning method of defect sector and access method for defect sector therefor}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a defect management method of a hard disk drive, and more particularly, to a method for dealing with a defect sector generated in a disk that is being used by a user, and a method for accessing a defect sector suitable thereto.

A disk drive is a magnetic recording device used for storing information, and has a form in which a plurality of concentric disks (commonly called media) are stacked. The information is recorded in concentric tracks formed on the surface of the disc. In order to facilitate data storage and retrieval in a regular manner, tracks on a disc are partitioned into blocks called sectors. The positional information of these sectors is represented by unique identifiers called cylinders (or tracks), heads (devices for accessing each surface of the disc) and sector numbers.

The disk is mounted to the spin motor and the information is accessed by read / write heads mounted to the actuator arm.

On the other hand, there is a defect portion in the disc which cannot read or write information. The disc may have its own defects, or defects may occur during production of the product or in use by the user.

In the process of assembling the actuator assembly to the media during the assembling of the hard disk drive, defects may occur on the media by scratching.

The physical form of damage caused by scratching results in loss of bytes, the basic unit of reading and writing data. On the other hand, since the controllable basic units of the disk drive are sectors and tracks, the loss due to scratches is at least one sector.

Such a bad sector (defected sector) becomes unusable when writing or reading data. Therefore, in the disc drive production process, the defect is found and replaced with the normal area, thereby providing the user with a flawless disc.

One production process for disk drives is the burn-in test. Burn-in tests are performed by repeatedly performing a recording / reading test while changing data patterns and recording / reading conditions to determine the presence and location of defects. Defect locations (cylinders, heads, sectors) and error codes identified through burn-in tests are stored in a log-in list to prevent users from accessing them later.

Defects can, of course, also occur during use by the user after the disk drive is shipped.

Despite the physical discontinuity of the sector due to defects, the sector address for accessing the sector from the outside must be contiguous. Therefore, the defect sector needs to be replaced with a normal sector in the spare area. Such a process is called a defect process.

There are various methods of defect processing, but there are slip methods and reassign methods.

The sleep method is mainly used to deal with defects detected in the burn-in test, and the reassignment method is mainly used to deal with defects generated during use (hereinafter referred to as reallocation defects).

The sleep method and the reallocation method are common in that they use a spare area, but sectors in which the sleep method replaces a defective sector by a contiguous sector and are shorter than the defective sector in the data area are sectors in the spare area. The reassignment method differs from the replacement method in that the defective sector is simply replaced with a sector in the spare area.

The sleep method has a high data access rate because the defected sector is replaced by the next consecutive sector, but the reassignment method has a slow data access rate because the defected sector is replaced by the sector in the spare area.

In recent years, disk drivers have become increasingly high in capacity, and as a result, the loss of defects as well as the capacity of reassignable defects are gradually increased to minimize data loss.

However, unlike the sleep method, in the reallocation method, as the number of reallocated sectors increases, more frequent sector searches occur during recording / reading, thereby causing processing delays.

Moreover, in the multimedia environment that is emerging as a new application area of the disk driver, such processing delay becomes an important factor enough to generate a claim.

SUMMARY OF THE INVENTION The present invention is designed to meet the above requirements, and an object thereof is to provide a new defect sector reallocation method capable of minimizing sector search and preventing data processing delay in processing reallocation defects generated during use. It is done.

Another object of the present invention is to provide a defect sector access method suitable for the defect sector reallocation method described above.

Another object of the present invention is to provide an improved defect sector list generation method.

1 is intended to schematically show how to use the spare area.

2 is shown to schematically show a slip scheme.

3 is a diagram illustrating a reallocation scheme using zone-specific spares.

Figure 4 is shown to schematically illustrate the reassignment scheme using per-cylinder spares.

FIG. 5 is a schematic view showing that re-rotation occurs in the reassignment method using the per-cylinder spare area.

6 is shown to schematically illustrate the role of sector pulses in a disk drive.

7 is a diagram schematically showing an operation of referring to sector pulse information of a slip list.

8 is a waveform diagram for schematically showing a read / write operation before performing reallocation.

9 is a flowchart illustrating a defect sector reallocation method according to the present invention.

10 is a waveform diagram for schematically showing an operation after the reassignment process.

11 is a flowchart illustrating an access optimization method of reallocation sectors according to the present invention.

Defect sector reallocation method according to the present invention to achieve the above object

In the method of processing a defect sector in a headerless ID hard disk drive,

If a defect sector occurs while the hard disk drive is in use, reading from the defect sector to the end of the corresponding track and storing in a buffer;

Storing contents stored in the buffer sequentially from the next sector of the defect sector to the spare area; And

And generating a reassignment list having sector pulse information from which the sector pulses of the defect sector have been removed.

The defect sector access method according to the present invention to achieve the above another object

If a defect sector occurs while using a headerless ID hard disk drive, the defect sector is read from the defect sector to the end of the corresponding track and stored in the buffer, and the contents stored in the buffer are sequentially stored from the next sector of the defect sector. Creating a reallocation defect list having sector pulse information from which sector pulses of the defect sector are removed;

Acquiring sector pulse information from a sleep defect list having sector pulse information from which sector pulses of a defect sector are removed before shipment of the hard disk drive;

Updating sector pulse information obtained from the sleep defect list by sector pulse information of the reassignment defect list; And

And accessing a defect sector by the updated sector pulse information.

Defect list creation method according to the present invention for achieving the above another object

In the method for creating a defect list having information related to a defect sector in a headerless ID hard disk driver,

Creating a reassignment defect list having sector pulse information from which sector pulses of defect sectors generated during use of the hard disk drive are removed;

Acquiring sector pulse information from a sleep defect list having sector pulse information from which a sector pulse of a defect sector generated before shipment of the hard disk drive is removed;

Obtaining sector pulse information from the reallocation defect list; And

And updating the sector pulse information obtained from the sleep defect list by the sector pulse information obtained from the reassignment defect list.

Hereinafter, the configuration and operation of the present invention will be described in detail with reference to the accompanying drawings.

1 is intended to schematically show how to use the spare area.

The spare area is used in one of two ways. One is a zone-specific spare area as shown in FIG. 1A, and the other is a cylinder-specific (or track) spare area as shown in FIG. 1B.

The zone-specific spare area divides the concentric circular tracks into several areas (zones) in the radial direction and has a spare area as shown by hatched portions in FIG. 1A for each zone. Therefore, the spare area is composed of a plurality of tracks, and is set as the tracks on the innermost side in each zone for ease of access.

Spare area per cylinder is to have spare area as shown by hatched in FIG. 1B per cylinder. Therefore, the spare area is composed of a plurality of sectors and is set for each track.

On the other hand, there are two types of defect processing methods. One is the slip method and the other is the reallocation method.

According to the defect processing method and the spare area allocation method, there are three defect processing methods for each spare area. The sleep method is applicable only to the spare area for each cylinder because the spare area requires that the spare area be adjacent to the data area.

Accordingly, there is a slip method using a spare area for each cylinder, a reallocation method using a spare area for each zone, and a reallocation method using a spare area for each cylinder.

The slip method using a spare cylinder area (hereinafter, simply referred to as a sleep method) slips an address one by one from the sector immediately following the defect sector to the spare area when the defect occurs so that the defective sector is not accessed.

The reallocation method using zone-specific spare areas divides the drive into multiple zones in the radial direction, each zone has a spare area (spare cylinder), and replaces the defective sector with a spare sector located in the spare cylinder when a defect occurs. So that the defect sector is not accessed.

In the reallocation method using a spare area for each cylinder, a spare area is provided for each cylinder, and when a defect occurs, the defective sector is replaced with a spare sector located in the spare area so that the defective sector is not accessed.

Figure 2 is shown to show a slip scheme schematically, the upper side shows the state before the slip occurs, the lower side shows the state after the slip occurs.

On the upper side of Fig. 2, the tracks, which are actually concentric circles, are shown in a straight line, D0-D701 representing data sectors, and S0-S8 representing spare sectors. In addition, successive serial numbers indicate that the sectors are immediately adjacent to each other.

In the upper side of Fig. 2, the defective sector D1 of the hatched portion should be replaced by the spare sector of the spare area. In the sleep method, the defect sector D1 is replaced by the following process.

1) Defected sector D1 is replaced by adjacent sector D2, and an address of D1 is assigned to D2.

2) D2 is again replaced by an adjacent sector D3, and an address of D2 is assigned to D3. This approach is performed for all sectors of the data area.

3) As a result of the substitution by adjacent sectors in the data area, at the end of the data area there are no more sectors to allocate. Accordingly, the last sector D701 of the data area is replaced by the first sector S1 of the spare area.

The lower side of FIG. 2 shows a state after the slip using the spare area for each cylinder occurs. In the lower part of Fig. 2, the sectors marked with black blocks in the data area means that they are no longer used as defected sectors. Further, the arrow means that the defective sector D1 is replaced by the adjacent sector D2, and D2 is accessed by the address of D1.

In the sleep method, even when there are defect sectors, since the sectors are continuously positioned, all sectors can be accessed by one seek operation.

3 is a diagram illustrating a reallocation scheme using zone-specific spares.

Sector D2, indicated by black blocks in the data area in FIG. 3, means that it is a decoded sector and that it is no longer used. Further, the arrow means that the defected sector D2 is replaced by the sector S1 of the spare area.

Normally, spare cylinders are placed at the end, or innermost, end of each zone, and if any sector in that zone is defective, the spare cylinder is replaced with the sectors that are replaced by reallocation when accessing the defective sector. The sector of the spare cylinder is accessed.

In recent years, disk drives have grown in capacity to reach thousands of cylinders in a zone. If reallocation is performed on the first track in a zone with 3000 cylinders (tracks), that is, if there is a defect sector on the first track and it is replaced by a spare sector in the spare cylinder, each time the defective sector is accessed The drive seeks for 3000 tracks each time. seek refers to a track search, in which case a seek for searching a spare cylinder to access a defected sector and a seek for returning back to the original track are performed.

In the worst case, when reallocation occurs across one sector, the transfer speed is significantly slowed at a specific location since 3000 tracks must be accessed every sector.

Figure 4 is shown to schematically illustrate the reassignment scheme using per-cylinder spares. In FIG. 4, the tracks, which are actually concentric circles, are shown in a straight line, D0-D703 representing data sectors, and SPARE1 and SPARE2 represent spare sectors. In addition, successive serial numbers indicate that the sectors are immediately adjacent to each other.

Sector D2, indicated by black blocks in the data area in FIG. 4, means that it is a decoded sector and is no longer used. Further, the arrow means that the defected sector D2 is replaced by the sector SPARE1 of the spare area.

In the reassignment scheme using per-cylinder spare sectors, track seek does not occur, but three revolutions are required for access because the spare sector is behind the cylinder. Here, rerotation means that the disc is rotated once to search for tracks located in the same track.

FIG. 5 is a schematic view showing that re-rotation occurs in the reassignment method using the per-cylinder spare area. In FIG. 5, the upper figure shows that a defect has occurred in the data block D2 (black block), and the lower figure shows that the defective data block D2 is replaced by the spatter sector SPARE 1. FIG.

First, a first rerotation is performed to access the data sector D2, then a second rerotation is performed to access the spare sector SPARE1, and finally a third rerotation is performed to access the next data sector D3.

As described with reference to FIGS. 3 to 5, in the conventional reallocation scheme, an additional seek (zone-specific spare cylinder) is performed because the position of a sector (alternative sector alternate sector) to replace the defect sector is not continuous with the defective data sector. Is used) and re-rotation (when using a cylinder-specific spare sector).

The delay caused by additional seek and re-rotation can have a fatal effect in multimedia environments using A / V (Audi / Video) data.

In the present invention, when a defect to be reassigned occurs, the defective sector is read from the last sector of the track and then rewritten from the next sector of the defect sector, and the remaining sector at the end of the track is recorded in the spare sector. Thereby minimizing the delay incurred when accessing the defect sector.

6 is shown to schematically illustrate the role of sector pulses in a disk drive.

The figure on the upper side of FIG. 6 shows that the tracks, which are actually concentric circles, unfold in a straight line. D0-D701 are data sectors, and spare1 and spare2 are spare sectors.

Each sector is written and read out in synchronization with a sector pulse. Sector pulses are stored in a slip list that records slip information and are referenced in write and read operations.

The waveform diagram in the center of FIG. 6 shows a sector pulse when there are no defects in the track, and the waveform diagram below in FIG. 6 shows a sector pulse when there are defects in the track (D2 in the figure). Seems. Comparing the middle and lower waveforms of FIG. 6, it can be seen that sector pulses do not occur in the defected sector D2 as compared with sector sectors in the normal sector. Here, sector pulse information referred to to generate sector pulses is in a slip list.

7 is a diagram schematically showing an operation of referring to sector pulse information of a slip list. When the disk drive is initialized, the microprocessor 72 reads the slip list 74 and loads it into the internal buffer 76. The slip list 74 is recorded in a maintenance cylinder which is not accessible by the user.

In accessing each track, the microprocessor reads the sector pulse information of the target track from the buffer 76 and loads it into the internal SRAM 78 of the sequencer. The sequencer refers to an apparatus for generating sector pulses per track. The sequencer generates sector pulses at the head of each sector as shown in the center waveform diagram of FIG. 6 with reference to sector pulse information loaded into its internal SRAM 78. If there is a defected sector, it will not generate a sector pulse as shown in the lower waveform diagram of FIG. Accordingly, the writing and reading of the sector is performed in synchronization with the sector pulse, and since the sector pulse does not occur in the defective sector, the writing and reading operation is not performed in the defective sector.

On the other hand, when a seek is performed like a disk write / read (line write and read) operation, the slip list always checks whether there is slip on the target track, and then loads the contents into the internal SRAM 76 of the sequencer.

The operation of the present invention will be described after dividing into before and after reassignment.

1) Operation before reassignment

(1) read / writ operation

There can be various causes of reallocation such as read error and write error. Among them, make sure that there is no problem in rewriting on the track as in the case of read error. If there is no problem in rewriting the corresponding track, the decoded sector is read and stored in the buffer from the last sector of the track to which the defected sector belongs. The contents stored in the buffer are recorded in order from the next sector of the defect sector. The contents of the defect sector are thus recorded in the next sector of the defect sector, and this is repeated until the last sector. Here, the contents of the last sector stored in the buffer are recorded in the spare sector of the spare area.

In the headerless ID method, the current sector number is calculated by counting the number of sector pulses from the index, and the sequencer performs the read / write operation as described above from the sector pulses at the time when the defective sector is encountered during the write / read operation. do.

(2) reassign list creation operation

When a disk drive seeks through a write / read operation, it always checks whether there is a slip in the target sector in the slip list. The contents are then loaded into the sequencer's internal SRAM 76 for use in generating sector pulses.

The slip list contains information about defects that existed before the disk drive was shipped, but does not contain information about defects that occurred while the disk drive was in use.

Therefore, if sector pulse information related to the defective sector can be modified after slip list loading, sector pulse timing can be freely adjusted without modifying the slip list.

That is, after recording the contents of the last sector of the track from the defected sector as in (1) from the next sector to the spare sector of the defected sector, the sector pulse is not generated in the defected sector. Modifying the information allows access to reallocated defect sectors without seek and re-rotation as in the slip method.

However, the slip list is stored in the management cylinder and cannot be updated after the disk drive is shipped.

According to the present invention, a reassignment list having sector pulse information obtained by removing sector pulses of a defect sector generated during use is provided, and a sector pulse is referred to with reference to a slip list and a reassignment list when a track including a defective sector is accessed. To generate.

That is, in generating sector pulses, not only the slip list but also the reassignment list are referred to.

8 is a waveform diagram for schematically showing a read / write operation before performing reallocation.

FIG. 8A shows that reallocation defects have occurred in the data sector D2 (black block).

8B shows reading from the data sector D2 where the reallocation defect occurred to the last sector of the track. Here, RG / WG (ReadGate / WriteGate) is a gate signal that controls read / write operations. The contents from the defected data sector D2 to the last sector of the track are stored in the buffer.

FIG. 8C shows that the contents stored in the buffer are recorded from the next sector D3 to the spare sector spare1 of the defective data sector D2.

As a result of such a recording operation, as shown in the upper side of Fig. 8C, contents from the defect-produced data sector D2 to the last sector are recorded by one sector.

9 is a flowchart showing a defect processing method according to the present invention.

First, a target sector is read (s902).

It is checked whether there is an error in the target sector (s904). If there is no error, the next reallocation defect processing operation is skipped and the next operation, for example, the operation of reading the next sector is performed.

If there is an error, check whether the data is recoverable (s906).

If the data is recoverable, it is checked whether the next sectors, that is, from the next sector to the spare sector of the target sector, are writable.

If the data is not recoverable or not writable from the next sector to the spare sector of the target sector, the process branches to the error code process (s918) (s908).

If the data is recoverable and recordable from the next sector to the spare sector of the target sector, it is determined that a defect has occurred and the data is read from the target sector to the last sector. The read data is stored in the buffer (s910).

The sector pulse information read from the sleep list is corrected. Sector pulse information read from the sleep list is stored in the internal RAM of the disk drive and the internal RAM of the sequencer. Among them, the sector pulse of the defect sector is erased from the sector pulse information stored in the internal RAM (s912).

Data stored in the buffer, that is, data from the target sector to the last sector, are recorded from the next sector to the spare sector of the defect sector (s914).

The sector pulse information created in step s912 is recorded in the reassignment list (s916).

The reassignment list to be referred to in the read / write operation after the reallocation of the result of s912 is created.

2) Actions after reassignment

As mentioned above, during the write / read operation, sector pulse information is first stored in the sequencer's RAM by referring to the slip list.

Next, the reallocation list is read to remove sector pulses of the defect sectors from the sector pulse information stored in RAM.

As a result, the defective sector is no longer accessed when read or written.

10 is a waveform diagram for schematically showing an operation after the reassignment process.

FIG. 10A shows a sector pulse generated by sector pulse information read from a sleep list loaded in an internal RAM of a disk drive. As shown in Fig. 10A, the sector pulse generated by the slip list also occurs in the defect sector. This is because the slip list does not have sector pulse information for the reallocated defect vector.

10B illustrates sector pulses generated by referring to the reassignment list. As compared with FIG. 10A, it can be seen that no sector pulse is generated at the defect sector D2. This is the result of modifying the sector pulse as shown in Fig. 10A with reference to the sector pulse information recorded in the reassignment list.

11 is a flowchart illustrating a method of accessing a reallocation sector according to the present invention.

First, a read / write command is generated (s1102).

The target track is searched (s1104).

When the target track is found, the slip list is read (s1106).

The sector pulse information is loaded into the sequencer's internal RAM (s1008). That is, the previous sector pulse information stored in the sequencer's internal RAM is deleted, and the sector pulse information suitable for the target track is loaded (s1108).

Read the reallocation list (s1110).

The sector pulse information stored in the internal SRAM 76 of the sequencer is corrected with reference to the colorant pulse information of the reassignment list. That is, the sector pulse corresponding to the reallocated defect sector is erased. Thereafter, the defect sector is accessed by the updated sector pulse information (s1012).

Defect list creation method according to the present invention

A process of creating a reallocation defect list and updating a sleep defect list by the created reallocation defect list, the actual contents of which are disclosed in the detection sector access method described with reference to FIG. Will be omitted.

First, create a reallocation defect list.

Next, sector pulse information is obtained from the sleep defect list. The sleep defect list contains sector pulse information in which the sector pulses of the defect sectors generated before the shipment of the hard disk are removed.

Next, sector pulse information is obtained from the reallocation defect list. The reassignment defect list has sector pulse information from which sector pulses of the defect sectors reallocated by the defect sector reallocation method of the present invention are removed.

Finally, the sector pulse information obtained from the sleep defect list is updated by the sector pulse information obtained from the reallocation defect list.

That is, the sector pulse corresponding to the reallocated defect sector is erased from the sector pulse information obtained from the sleep defect list.

According to the defect sector reallocation method according to the present invention as described above, since access to the reallocated defect sector is not required for track search and re-rotation, processing delay of data can be prevented.

In addition, when the defect sector reallocation method of the present invention is applied to recording / storing a video using a hard disk drive in a personal video recorder (PVR) or the like, screen interruption may be prevented due to data processing delay due to defects. Can be.

In addition, even when data is recorded on a DVDR / CD-R or the like, original data is often stored in a hard disk drive, but a recording error may occur when the data stream from the hard disk drive is not constant. By applying the present invention, it is possible to prevent such data streams from being inconsistent, thereby preventing writing errors.

As described above, the reassignment defect processing method according to the present invention has an effect of minimizing the execution speed reduction due to the reassignment defect.

In addition, when storing and reading a video from a recent PVR (Personal Video Recorder) such as a hard disk drive (HDD), there is an advantage that can be prevented from screen interruption due to reassignment defect.

Furthermore, when data is recorded by an optical recording medium such as DVDR / CD-R, original data is usually recorded on the HDD, which reduces the recording error that occurs when the data stream from the HDD is not constant. .

Claims (3)

In the method for processing a defect sector in a headerless ID hard disk drive, If a defect sector occurs while the hard disk drive is in use, reading from the defect sector to the end of the corresponding track and storing in a buffer; Storing the contents stored in the buffer sequentially from the next sector of the defective sector to the spare area; And And creating a reassignment list having sector pulse information from which the sector pulses of the defect sectors have been removed. If a defect sector occurs while using a headerless ID hard disk drive, the defect sector is read from the defect sector to the end of the corresponding track and stored in the buffer, and the contents stored in the buffer are sequentially stored from the next sector of the defect sector. Generating a reassignment list having sector pulse information from which sector pulses of the defect sector have been removed; Acquiring sector pulse information from a sleep defect list having sector pulse information from which sector pulses of a defect sector are removed before shipment of the hard disk drive; Updating sector pulse information obtained from the sleep defect list by sector pulse information of the reassignment defect list; And And accessing a defect sector by the updated sector pulse information. In the method for creating a defect list having information related to a defect sector in a headerless ID hard disk driver, Creating a reassignment defect list having sector pulse information from which sector pulses of defect sectors generated during use of the hard disk drive are removed; Acquiring sector pulse information from a sleep defect list having sector pulse information from which a sector pulse of a defect sector generated before shipment of the hard disk drive is removed; Obtaining sector pulse information from the reallocation defect list; And And updating the sector pulse information obtained from the sleep defect list by the sector pulse information obtained from the reassignment defect list.