JP2001051807A - Defective storage block substituting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To complete access to a segment in a determined time even after a defective storage block is substituted for by selecting one or at least two storage blocks as alternative areas from the standby recording area according to an allocation state of the alternative areas of the segment and replacing the storage blocks including the defective storage block. SOLUTION: A control part 102 at the time of detecting a new defect registers the storage block having the defect as the defective block. When there are >2 registered defects in a continuous recording area that the newly detected defect belongs to, the control part 102 selects two defective blocks included in the continuous recording area that the newly detected defect belongs to. Then the control part 102 allows continuous access to two selected defective blocks generated after the selection and storage blocks arranged successively between them by replacing the access with access to the successively arranged storage blocks in the standby storage area.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for replacing a defective storage block in a storage device such as a disk device comprising a plurality of storage blocks, and more particularly to a defect for preventing a decrease in data read / write performance after defective block replacement processing. It relates to a block replacement device.

[0002]

2. Description of the Related Art Conventionally, as a defective block replacement device for a storage device composed of a plurality of storage blocks, a method of replacing a defective storage block with a replacement storage block is known. To describe this more specifically, in a storage device such as a disk device, the entire storage area is configured by storage blocks called a plurality of sectors. When a defective storage block, that is, a defective sector occurs, a method of searching for a replacement sector for the defective sector and replacing the defective sector in sector units is used. Also,
For the selection of a replacement sector for a defective sector, a method of selecting a replacement sector closest to the physical position of the defective sector is used.

However, in a storage device composed of a plurality of storage blocks, normally, when reading and writing are successively performed on adjacent (or adjacent) storage blocks, distant storage blocks are sequentially read. Higher performance can be obtained compared to writing. For this reason, as shown in FIG. 21, when the data storage destination of a certain storage area A becomes discrete as a result of the substitution in units of storage blocks (sectors), the performance of the read / write execution of the entire storage area A is significantly reduced. A phenomenon occurs.

[0004] As a defective storage block replacement device for preventing a decrease in data read / write performance after such a defective block replacement process, there is known a method described in a disk sector replacement device of JP-A-4-141721. In this method, when a bad sector occurs, the entire area composed of a plurality of sectors called a record is replaced with another record. By substituting on a record-by-record basis, it is possible to prevent the occurrence of a large number of substitutive sectors in a record as shown in FIG. 21 and a decrease in the read / write performance of the entire record.

Further, Japanese Patent No. 2591427 (Japanese Patent Laid-Open No. 6-33381) discloses a method of managing addresses of defective storage blocks.
No. 40) is known. FIG. 20 shows a configuration example of a conventional address management method for a defective storage block, in which a memory 201 for storing a defect position, an address generation 202 for generating an address for the memory 201, and an address generation 202 for the address generation 202. And a counter 204 for outputting position data on the magnetic disk. The counter 204 counts up the address to inspect the magnetic disk, notifies the address to the magnetic disk, and also notifies the memory 201 and the control unit 203. When detecting a defect, the control unit 203 writes data in the memory 201. The memory 201 writes the write signal and the position data from the counter 204 to the address of the address generation 202 when the write signal is given. After writing, the control unit 203 checks the value from the counter 204 and, when the predetermined value is counted up, instructs the address generation 202 to update. That is, a defect detected before updating the address is registered as one defect by writing to the same address in the memory 201. The data of the area registered as a defect is replaced with a vacant area of the alternative area.

[0006]

When handling real-time data such as video, read / write (hereinafter, referred to as access) is executed in units of blocks larger than the storage block of the storage device. A large block, which is a unit for accessing the storage device, is hereinafter referred to as a segment. Taking video data as an example, usually, a segment has a data size of one frame of video. This depends on the resolution of the video and the compression method, but is about 128 KB to 1 MB. When handling real-time data such as video data,
Access to this segment must be completed within a predetermined time. If the access to the segment is not completed within a predetermined time, it may result in an obstacle such as a disturbed image. For this reason, in a replacement device for a defective storage block of a storage device that stores real-time data, it is demanded that access to a segment is always completed within a predetermined time.

However, in the conventional method of substituting storage blocks (sectors) in units, when a large number of storage blocks in a segment are subjected to substitution processing, actual storage of data is performed as shown in FIG. There is a problem that the destination becomes discrete, and as a result, access to the segment cannot be completed within a predetermined time. In addition, if the method of selecting the replacement sector closest to the physical position of the defective sector as the replacement sector of the defective sector is used, depending on the physical arrangement of the sectors located before and after the logical address of the defective sector, the segment may be formed as a segment. At the time of access, there is a problem that a large delay time may occur due to substitution of a sector.

Japanese Patent Laid-Open No. 6-338140 describes a method of managing the address of a storage block in which a defect has occurred. It is not stated whether a block is allocated. Therefore, when it is considered that a replacement block is allocated in units of a storage block which has been conventionally used, as described above, a problem occurs that access to a segment cannot be completed within a predetermined time. More specifically, the method described in the magnetic disk device of Japanese Patent Application Laid-Open No. 6-338140 manages a defective storage block generated in close proximity as one defective area,
This patent discloses a method for saving the amount of memory for managing addresses of defective storage blocks. If a large number of defective areas are generated in a certain storage area (segment), a data storage destination is determined in the same manner as in the previous case. Becomes discrete, and as a result, a problem arises in that access to a segment cannot be completed within a predetermined time. In other words, the method disclosed in the magnetic disk device of Japanese Patent Application Laid-Open No. 6-338140 cannot essentially solve the problem that access to a segment cannot be completed within a predetermined time.

On the other hand, according to the method described in the disk sector replacement device of Japanese Patent Application Laid-Open No. H4-141721, when a record is equal to a segment, it is possible to avoid a data storage destination being discrete, Access can be completed within a predetermined time. However, in the method described in the disk sector replacement device of JP-A-4-141721, a storage area for replacement of a segment size is required every time a defect occurs in one storage block. When the segment size is large, there is a problem that a large number of replacement storage blocks are required.

The present invention solves such a conventional problem, efficiently uses a replacement storage block, and completes access to a segment within a predetermined time after replacement of a defective storage block. It is an object of the present invention to provide an alternative device for a defective storage block that can be made to be defective.

[0011]

SUMMARY OF THE INVENTION In order to solve this problem, an invention according to claim 1 comprises a storage block of a minimum recording unit of a recording device that can be randomly accessed and a plurality of the storage blocks, In, the access order of the storage block is continuous on the recording device, has a segment of a unit for accessing data, and a spare recording area managed in units of an area in which the storage block is continuously arranged, Detecting means for detecting the storage block having lost the initial function in access as a defective storage block, and detecting the defective storage block,
Selecting means for selecting the storage block from the spare recording area as an alternative area; and accessing the storage block constituting the segment, when accessing the storage block to which the alternative area is assigned, the storage block of the alternative area. A defective block replacement device having access address control means for controlling an access position so as to perform an access of the segment, wherein the selection means is configured to store one piece of data from the spare recording area based on a state of assignment of a replacement area of the segment. A block is selected as a replacement area, or at least two storage blocks are selected as replacement areas from the preliminary recording area to replace a plurality of storage blocks including the defective storage block.

[0012] In this way, when the replacement is performed in units of storage areas (segments), it is guaranteed that the access to the segment is completed within a predetermined number of times after the replacement processing.

According to a second aspect of the present invention, in the first aspect, the alternative area selecting means detects one storage block from the spare recording area when the number of the alternative areas is less than N in the segment. The selected defective storage block is selected as a replacement area, and the number of replacement areas in the segment is N.
When the storage blocks are arranged in the order in which the reading process is executed in the segment, the adjacent replacement area including the detected defective storage block or the replacement area at both ends or the detected replacement area is used. Among the N sections surrounded by the defective storage blocks, the section having the smallest number of storage blocks existing in the section is detected, and all the storage blocks existing in the section having the smallest number of storage blocks are replaced. An alternative area is selected from an area continuously arranged in the preliminary recording area, and when the detected defective storage block is not included in the section having the smallest number of storage blocks, one area is selected from the preliminary recording area. The storage block is selected as a replacement area for replacing the detected defective storage block.

Thus, in the case where the replacement is performed in units of storage areas (segments), it is assured that the access to the segment is completed within the predetermined number of times after the replacement processing with a smaller number of replacement areas.

According to a third aspect of the present invention, in accordance with the first aspect, the alternative area selecting means detects one storage block from the spare recording area when the number of the alternative areas is less than N in the segment. The selected defective storage block is selected as a replacement area, and the number of replacement areas in the segment is N.
If the number is the same, the storage blocks are arranged in the order in which the reading process is executed in the segment, and when considering the annular region where the first storage block and the last storage block of the segment are adjacent to each other, it is detected that both ends are located. Among the N + 1 sections surrounded by the defective storage block or the adjacent replacement area including the replacement area or the detected storage block, the section having the smallest number of storage blocks in the section is detected. A replacement area for replacing all the storage blocks existing in the section having the smallest number of storage blocks is selected from an area in which the spare recording area is continuously arranged, and the number of storage blocks in which the detected defective storage block is smaller than the number of storage blocks is selected. When not included in the least said section,
One storage block is selected from the spare recording area as a replacement area for replacing the detected defective storage block.

Thus, when the replacement is performed in units of storage areas (segments), it is guaranteed that the access to the segment is completed within a predetermined number of times after the replacement processing with a smaller number of replacement areas.

According to a fourth aspect of the present invention, in accordance with the first aspect, the alternative area selecting means removes one storage block from the spare recording area when the number of the alternative areas is less than N in the segment. When a storage block is selected as a replacement area and the number of replacement areas in the segment is N, both ends are obtained when the storage blocks are arranged in the order in which the reading process is executed in the segment. Surrounded by the first storage block or last storage block of the segment or the first storage block or last storage block of the adjacent segment including the detected defective storage block or replacement area or the replacement area or the detected defective storage block Among the N + 2 sections, the section having the smallest number of storage blocks existing in the section is detected, and the storage block is detected. The replacement area for replacing the entire storage block click number is present in the smallest within the interval, along with selected, taking consecutive arranged in the region of the preliminary recording area,
When the detected defective storage block is not included in the section having the smallest number of storage blocks, one storage block is selected from the preliminary recording area as an alternative area for replacing the detected defective storage block. Features.

Thus, when the replacement is performed in units of storage areas (segments), it is guaranteed that the access to the segment is completed within a predetermined number of times after the replacement processing with a smaller number of replacement areas.

The invention according to claim 5 is the invention according to claims 2 to 4.
, The value of N is set to 2.

Thus, when the replacement is performed in units of storage areas (segments), it is guaranteed that the access to the segment is completed within one time after the replacement processing.

According to a sixth aspect of the present invention, in accordance with the first aspect of the present invention, the alternative area selecting means is arranged so that the number of the alternative areas in the segment is one and the storage blocks are arranged in the order in which the reading process is executed in the segment. Are aligned, including the replacement area and the defective storage block, a replacement area that replaces all the storage blocks included in the section surrounded by the defective storage block and the replacement area is continuously formed by the preliminary recording area. When the number of alternative areas in the segment is selected from the arrayed areas and the storage blocks are arranged in the order in which the reading process is executed in the segment, the first storage block or the last storage block of the segment And a section in which the number of storage blocks existing in the section is small among two sections surrounded by the detected defective storage block, An alternative area selecting means for selecting an alternative area that replaces all storage blocks existing in the detected section, including the lock or last storage block and the defective storage block, from an area in which the spare storage area is arranged continuously. It is characterized by having done.

Thus, when the replacement is performed in units of storage areas (segments), it is ensured that the access to the segment is completed within a predetermined smaller number of times after the replacement processing.

According to a seventh aspect of the present invention, in accordance with the first aspect, when the occurrence of a defective storage block is detected in a segment, the alternative area selecting means includes information on the detected defective storage block and information in the segment. One of the spare recording areas is selected as a replacement area for replacing the defective storage block, or the entire segment is replaced, using the information on the replacement area assigned to the storage block Alternative areas to
The spare area is selected from an area in which storage blocks are continuously arranged.

According to this, by using the information on the detected defective storage block and the information on the replacement area assigned to the storage block of the segment, the replacement device for the defective block is selected, and thereby the storage block is selected. If the access to the segment can be completed within the specified time after the replacement process even if the replacement method is used in units, if the replacement process is executed in storage block units and the access cannot be completed within the time, Control can be performed so that the substitution process is executed in segment units.

[0025] The invention of claim 8 is characterized in that, in the invention of claim 7, the replacement area for replacing the segment is an area in which reading and writing of the entire replacement area can be executed within a predetermined time. is there.

Thus, when the replacement is performed in units of storage areas (segments), it is guaranteed that the access to the segment is completed within a predetermined time after the replacement processing.

According to a ninth aspect of the present invention, in accordance with the seventh aspect, the alternative area selecting means is configured such that no alternative storage block is assigned to the storage block in the segment, or the entire segment is replaced by the alternative area. If the number of defects in the area is zero, a replacement storage block is assigned if a replacement storage block is assigned to a storage block in a replacement area in which a segment or the entire segment is replaced. The number of stored memory blocks is defined as the number of defects in the area, and when the occurrence of a defective memory block is detected in the segment, the number of defects in the area is less than a predetermined threshold value or the number of detected defective memory blocks is one.
When the replacement area is composed of a plurality of storage blocks, one storage block is selected from the spare area as a replacement area for replacing only the detected defective storage block, and the number of defects in the area is equal to or greater than a predetermined threshold. Further, when the detected defective storage block is not a replacement area consisting of one storage block, a replacement area for replacing the entire segment is selected from the area in which the storage blocks of the spare area are continuously arranged. And This makes it possible to easily select an alternative device.

According to a tenth aspect of the present invention, in accordance with the seventh aspect, when the occurrence of a defective storage block is detected in the segment, the alternative area selecting means sets one of the spare areas for the defective storage block. After temporarily selecting a storage block as a substitute area candidate, a delay evaluation value which is an evaluation value of an access time is obtained by using address information of each storage block in which data is actually recorded in the segment, and the delay evaluation value is calculated. When the value is less than a predetermined threshold, the alternative area candidate is selected as an alternative area for replacing only the detected defect storage block. When the number of defects in the area is equal to or more than a predetermined threshold, the spare area is selected. Is selected from the area in which the storage blocks are continuously arranged.

Thus, it is possible to evaluate a change in access time due to the physical arrangement of each storage block used for replacement, and it is possible to precisely select a replacement device, and to provide a replacement storage block. Can be used more efficiently.

An eleventh aspect of the present invention is characterized in that the threshold value according to the ninth and tenth aspects is set based on an allowable execution time of reading and writing of the entire segment.

As a result, for example, as in a hard disk drive, the access time varies depending on the physical position of the storage area (usually, the storage area located on the outer circumference has a short access time, and the storage area located on the inner circumference has an access time). In the case where the time is long, the threshold can be set individually for each storage area. Further, even when the execution time in which access is permitted differs for each segment, it is possible to individually set each segment.

According to a twelfth aspect of the present invention, there is provided the defective storage block replacement device according to any one of the first to eleventh aspects,
The storage block that is not a defective storage block among the replaced storage blocks is the storage block that is normally used.

This makes it possible to effectively use the replaced recording block.

According to a thirteenth aspect of the present invention, in the defective storage block replacement device according to any one of the first to twelfth aspects, the storage block that is not a defective storage block among the replaced storage blocks is stored in the spare recording area. It is characterized by registering.

This makes it possible to use the replaced recording block effectively.

According to a fourteenth aspect of the present invention, in the defective storage block replacement device according to any one of the first to thirteenth aspects, the area in which the storage blocks are continuously arranged is replaced by the physical address or physical location of the storage block. Are adjacent or close to each other.

With this arrangement, high-speed access is possible in an area where the physical positions of the respective storage blocks are close to each other, for example, as in a hard disk device such as a hard disk drive. It is possible to provide an area that can be accessed at a high speed in a storage device that is slow.

According to the fifteenth aspect, when the occurrence of a defective storage block is detected, the logical address of the detected defective storage block is (A0).
The storage block of the spare storage area, which has the shortest execution time when the storage block of 1), the storage block of the logical address (A0), and the storage block of the logical address (A1 + 1) are sequentially executed, is referred to as the defective block. It is selected as a replacement recording block for the storage block.

By selecting this alternative storage block,
The access time when accessing a continuous logical address area including the logical address of the storage block subjected to the substitution processing can be reduced. As a result, it contributes to completing access to the segment within a predetermined time.

According to a sixteenth aspect of the present invention, when the alternative area selecting means according to any one of the first to fourteenth aspects selects, from the spare area, an alternative area for replacing a defective storage block. An alternative area is selected based on a method of selecting an alternative storage block.

This makes it possible to reduce the access time to the segment after the substitution processing.

According to a seventeenth aspect of the present invention, any one of the first to fourteenth and sixteenth aspects of the present invention is used for a magnetic disk.

As a result, it is possible to realize a magnetic disk in which access to a segment is guaranteed to be completed within a predetermined time even after execution of the substitution process.

[0044]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to FIGS.

(Embodiment 1) FIG. 1 shows blocks for explaining Embodiment 1 of the present invention. In FIG. 1, 101 is a disk device, 102 is a control unit, 103 is an address conversion unit, and 104 is an address conversion table. , 105 is a defect position management table, 106 is a spare area management table, 107
Is a disk.

The operation of the disk device configured as described above will be described below.

FIG. 2 is a diagram showing a specific segment taken out of a plurality of segments recorded in a continuous area.
FIG. 2 shows a schematic diagram of data recording on a general disk when real-time processing is required. For example, 1
One video frame is recorded as a segment. In FIG. 2, the segments are recorded from A10 to A19. A10 to A19 are recorded as continuous areas. Here, the continuous area is an area in which the overhead time (head seek, rotation waiting time) from the start of reading data is the shortest. FIG. 2 (1) shows a case where no defect has occurred in a specific segment described in this description.
FIG. 2B shows a case where one defect has occurred, and FIG. 2C shows a case where two defects have occurred. In (1) of FIG. 2, when no defect has occurred, when the host accesses the logical address segments A10 to A19 as segments, the physical addresses a10 to a19 on the disk are used.
Is converted to access. In (2) of FIG. 2, from a11 to a
12 shows the result of performing the substitution process with the 12 regions as defects. The data of the defective areas a11 to a12 are replaced with the substitute areas of b0 to b1. Further, (3) of FIG. 2 shows the result of performing the replacement process because the area from a21 to a22 becomes defective. The data in the area between a11 and a22 is
Replace with the continuous area from b1 to b2.

FIG. 3 shows an address conversion table 103, a defect position management table 105, and a spare area management table 106.
FIG. 7 is a diagram showing a specific change of a registered value. FIG. 3A shows a state before a defect is generated, and FIG.
(3) is a case where a second defect occurs in the segment.

The operation of FIG. 2 will be described with reference to the block diagram of FIG. 1 and the registered contents of each table of FIG.

First, when no defect has occurred, as shown in FIG. 3A, the size (A19-
The logical address in A10) is associated with the physical address from a10. If there is access from the host, the address conversion unit 103 performs address conversion with reference to the address conversion table 104. At this time, no defect has been detected, and no registration has been made in the defect position management table 105.
Reserved areas are registered in the spare area management table 106. Specifically, an area of size x from the physical address b0 is allocated.

Next, the case where the first defect occurs in the segment from A10 to A19 as shown in FIG. 2 (2) will be described. The control unit 102 detects that a defect has occurred between the physical addresses a11 and a12. Since the detected defect is the first defect in the segment including the detected defect,
B0 to b1 having the same size as to a12 are replaced with the replacement area. The control unit 102 performs registration in the address conversion table 104, the defect position management table 105, and the spare area management table 106. Thereafter, the address conversion unit 103 performs the conversion by referring to the address conversion table, so that the access from the host to the logical addresses A10 to A11 is converted from the physical addresses a10 to a11, and the logical addresses A11 to A11 are converted.
Access of 12 is converted from physical addresses b0 to b1, and access of logical addresses A12 to A19 is converted from physical addresses a12 to a19.
Convert to In this case, the address conversion table 104, the defect position management table 105, and the spare area management table 1
The registered value to 06 is as shown in (2) of FIG. For example, in the address conversion table 104, a logical address within a size (A11-A10) from A10 is registered so as to correspond to the leading physical address a10. Defect position management table 10
5 is registered as a defect from a11 to a12. The spare area management table 106 is used as an alternative area.
In the area other than the area between b0 and b1, the size (x-b1) from the start physical address b1 is reserved as a spare area.

Further, A10 to A19 as shown in FIG.
The operation performed when the second defect occurs in the segment described above will be described. The control unit 102 detects that a defect has occurred between a21 and a22. The control unit 102 is the second defect in the segment including the detected defect (here, the defect is determined by the number of defects, but may be determined by the number of replacement areas). And the area between them, b1
To b2 are assigned to the physical addresses of the alternative areas, and registered in the address conversion table 104, the defect position management table 105, and the spare area management table 106. Thereafter, the address conversion unit 103 performs the conversion by referring to the address conversion table, so that the access from the host to the logical addresses A10 to A11 is converted from the physical address a10 to a11, and the access from the logical addresses A11 to A22 is converted from the physical address b1. Converted to b2, access from logical addresses A22 to A19 is a physical address
Convert from a22 to a19. In this case, the registration values in the address conversion table 104, the defect position management table 105, and the spare area management table 106 are as shown in (3) of FIG. For example, in the address conversion table 104,
The logical addresses within the size (A11-A10) from A10 correspond to the leading physical address a10, and the logical addresses within the size (A22-A11) from A11 correspond to the physical addresses from b1. In the spare area management table 106, areas other than b1 and b2 are registered as usable areas. In the defect position management table 105, areas a11 to a12 and areas a21 to a22 are registered. Since the area between a12 and a21 is not registered as a defect position, it is allocated as an available area.

Further, A10 to A19 as shown in FIG.
The operation in the case where the third defect occurs in the segment No. will be described. The control unit 102 detects that a defect has occurred between a31 and a32. Since the control unit 102 is the third defect in the segment including the detected defect, the control unit 102 collects the three defects and the area between them and assigns b2 to b3 to the physical address of the replacement area. Since the registered values of each table can be registered in the same manner as in (3) of FIG. 3, the description is omitted. The same operation is performed after the fourth defect.

FIG. 4 shows an example of a flowchart when the control unit 102 performs the substitution process.

Although one specific segment has been described, when a defect occurs in another segment, it can be similarly operated by registering it in each table.

As described above, the number of accesses to the substitute area in one continuous segment can be limited.
Read one segment within a guaranteed fixed time,
Alternatively, an excellent defective block replacement device capable of performing writing can be realized.

(Embodiment 2) Embodiment 2 of the present invention
Is a disk device having the same configuration as in the first embodiment. The operation will be described below with reference to FIG. 1 used in the first embodiment.

FIG. 5 is a diagram showing one specific segment recorded in the continuous area. (1) and (2) in FIG.
Is the same as in the first embodiment. a10 to a12 area and a2
The areas from 1 to a19 are collectively replaced with a continuous area from b1 to b2 of the alternative area.

FIG. 6 is a diagram showing specific changes in registered values of the address conversion table 103, the defect position management table 105, and the spare area management table 106, as in the first embodiment. FIG. 5A shows a state before the occurrence of a defect, and FIG.
Is when one defect occurs in the segment,
(3) is a case where a second eye defect occurs in the segment.

The operation of FIG. 5 will be described with reference to the block diagram of FIG. 1 and the registered contents of each table of FIG. Until the second defect occurs, the operation is the same as that of the first embodiment, and the description is omitted. When the control unit 102 detects a defect between the physical addresses a11 and a12, which is the second defect, it operates as follows. The control unit 102 compares the combined size of the regions a10 and a12, the regions a21 and a19, and the sizes of the regions a11 and a22, and combines the smaller regions into one. Here, an example is shown in which the sizes of the areas a10 and a12 and the areas a21 and a19 are small. a10 to a12 area and a21
The replacement of the replacement area from b1 to b2 is performed by combining the areas from to to a19. At this time, the registered values of each table are as shown in (3) of FIG. For example, in the address conversion table 104, a logical address within the size (a12-a10) from A10 corresponds to the first physical address b1, and a size (a2
The logical address in 1-a12) corresponds to the head physical address a22, and the logical address in the size (a21-a12) from A22 is registered so as to correspond to the head physical address b1 + (a12-a10).
In the spare area management table 104, areas other than b1 and b2 are registered as usable areas. In the defect position management table 105, a11 to a12 and a21 to a22 are registered as defects. Since the area between a12 and a21 is not registered as a defect position, it is allocated as an available area.

Here, FIG. 7 shows an example of a flowchart when the controller 102 performs the substitution process.

Although the present embodiment has been described with respect to one specific segment, even when a defect occurs in another segment, the segment can be similarly operated by registering it in each table. When the sizes of the areas a11 and a22 are smaller, the same result as in the first embodiment is obtained. Here, when the second defect occurs in one segment, the control unit 102 performs the replacement process by combining the two defects into one. However, even when three or more defects occur, Applicable.

As described above, the number of accesses to the alternative area within one continuous segment can be limited, the size of the alternative area allocated can be reduced, and one segment can be allocated within a guaranteed fixed time. An excellent defective block replacement device that can perform reading or writing can be realized.

In the present embodiment, the areas at both ends are combined into one and the alternative area is allocated. However, it is needless to say that each area may be allocated to two alternative areas.

(Embodiment 3) Embodiment 3 of the present invention
Is a disk device having the same configuration as in the first embodiment. The operation will be described below with reference to FIG. 1 used in the first embodiment.

Since the operation is almost the same as that of the first embodiment, different points will be described. First, (1) and (2) in FIG. 8 are the same as in the first embodiment. FIG. 8C shows an example in which (N-1) defects occur in one segment and (N-1) defective blocks are registered. FIG. 8D shows a case where the Nth defect occurs. In this case, two defective blocks aN-11 and aN-1
2. Select aN1 and aN2, combine these two defective blocks and blocks between them into one, and replace them with a continuous area of the alternative area. The method of selecting two defective blocks may be, for example, selecting the closest defective block.
The method for registering each table is the same as that in the first embodiment, and a description thereof will be omitted.

Here, FIG. 9 shows an example of a flowchart when the controller 102 performs the substitution process.

As described above, the number of accesses to the replacement area in one continuous segment can be limited to N−1, and one segment can be read or written within a guaranteed fixed time. An excellent defective block replacement device that can be performed can be realized.

(Embodiment 4) Embodiment 4 of the present invention
Is a disk device having the same configuration as in the first embodiment. The operation will be described below with reference to FIG. 1 used in the first embodiment. Since the operation is almost the same as that of the second embodiment, the difference will be described. Embodiment 2
In the above, when there is a defect that is accessed two or more times in a segment, the area other than the defective block is collectively replaced. In the fourth embodiment, a defect that is accessed more than N times in the segment is detected. The difference is that in some cases, the area other than the defective block is collectively replaced. The method of registering each table is the same as in the second embodiment, and a description thereof will be omitted.

Here, FIG. 10 shows an example of a flowchart when the controller 102 performs the substitution process.

As described above, the number of accesses to the alternate area in one continuous segment can be limited to N-1 times, and one segment can be read or written within a guaranteed fixed time. This can realize an excellent defective block replacement device that can be used and can reduce the replacement area used in the third embodiment.

In the above description, an example has been described in which the control unit is provided inside the disk device. However, the present invention can be similarly implemented by providing a control unit on the host side.

(Embodiment 5) FIG. 17 shows the configuration of a disk drive according to Embodiment 5 of the present invention. 17, reference numeral 101 denotes a disk device, and 300 denotes a disk device 1.
01, a host device that issues read and write requests; 301, an address translation unit that translates a logical address into a physical address that uniquely indicates a storage area of a disk unit;
Reference numeral 302 denotes an alternative processing unit that selects and executes a sector unit alternative process and a segment unit alternative process, 303 denotes a control unit that accesses a disk unit in response to a request from the host device 300, and 304 denotes a disk that records data. Part, 30
Reference numeral 5 denotes an address conversion table indicating correspondence between logical addresses and physical addresses, and reference numeral 306 denotes a spare area management table for managing replacement segments and sectors.

The operation of the disk device configured as described above will be described below.

First, the relationship between the address specified by the host device 300 at the time of issuing the request and the segment, the disk unit 30
4 will be briefly described. The disk unit 304 is composed of a plurality of recording blocks (sectors), and a physical address is assigned to each sector. The disk unit 304 identifies a sector by a physical address and executes access (read / write). The host device 300 requests access to the disk device 101 at a logical address different from the physical address. One logical address indicates a storage block having a sector size.
The logical address is an address that does not change even when the replacement process is performed, and the host device 300 can access the disk device 101 without being aware of the replacement process. The host device 300 executes reading and writing on a segment basis, which is a block larger than a sector. Here, the segment size is an integral multiple of the sector size. Each segment is taken in an area where logical addresses are continuous. Host device 300
Sends an access request to the disk device 101 by designating the head logical address of the segment and the number of logical addresses constituting the segment.

Next, the operation at the time of receiving the access request from the host device 300 will be described. The control unit 3 of the disk device 101 that has received the access request from the host device 300
In step 03, the start logical address of the segment and the number of logical addresses constituting the segment are extracted from the access request.
Then, a logical address group forming the segment is created by sequentially incrementing the leading logical address taken out. The control unit 303 transmits the created logical address group to the address conversion unit 301. The address conversion unit 301 uses the address conversion table 305 to
The logical address is sequentially converted from the logical address to the physical address, and a physical address group is returned to the control unit 303. Upon receiving the physical address group, the control unit 303
The access is executed by designating a physical address group in 04. When the access to the disk unit 304 is completed,
The control unit 303 transmits the read data or the write completion response to the host device 300 according to the type of the request.

FIG. 18 shows a configuration example of the address conversion table. The address conversion table 305 shows correspondence between logical addresses and physical addresses. As the replacement flag in the address conversion table 305, “1” is set for the logical address on which the replacement process has been performed in sector units (the use of this replacement flag will be described later). For example, in the example of FIG. 18, the logical address “A000” is associated with the physical address “a000”. The logical address “A001” is initially associated with the physical address “a001”, but as a result of the sector-by-sector replacement processing, the physical address “i001”
The correspondence has been changed to "001".

Next, the operation of the substitution process will be described. FIG. 20 is a flowchart showing an example of the substitution process. Hereinafter, the operation of the substitution process will be described with reference to this flowchart.

The disk unit 304 finds a defective sector,
The physical address of the defective sector is notified to the control unit 303 (S101).

The control unit 303 transmits the physical address of the notified defective sector, the start logical block address and the last logical block address of the segment including the defective block to the substitution processing unit 302, and requests the substitution processing. The substitution processing unit 302 that has received the request determines the method of the substitution process based on the information notified from the control unit 303 and the information in the address management table (S102). The details of the method of selecting the alternative processing method will be described later.

When the alternative processing method for each sector is selected, the physical address of the alternative sector is obtained from the spare area management table 306 (S106), and the physical address indicated as a defective sector in the address management table 305 is obtained. Replaced with the physical address of the replacement sector. At this time, the replacement flag corresponding to the replaced physical address is set to "
1 "(S107).

If the replacement process is selected for each segment, a physical address group of the replacement segment is obtained from the spare area management table 306 (S103). Then, the physical address corresponding to the logical address of the segment specified by the control unit 303 of the address conversion table is replaced with the obtained physical address group of the replacement segment. At this time, all the substitution flags corresponding to the replaced physical address group are set to "0" (S105).

Further, of the physical addresses before replacement, those other than the physical address pointed out as a defect are registered in the spare area management table 306 as replacement sectors (S
104).

As described above, when performing replacement in segment units, by reusing the sector forming the segment before replacement as a replacement sector, the recording block can be effectively used, and the replacement area can be reduced. This makes it possible to configure a disk device.

Further, the replacement segment obtained for performing the replacement process in segment units is allocated from a physical area in which the access time to the segment can be shorter than the maximum access time to the segment permitted by the host device 300. By controlling in this way, it is guaranteed that the access can always be made within the maximum access time when the substitution process is executed in segment units. As an example of the substitute segment to be allocated, an area configured by continuous physical addresses, an area configured by adjacent physical addresses, and the like can be considered. When the allowable maximum access time is a very strict value, in addition to the above, it is conceivable that the physical area is located on the outer peripheral side. In order to allocate a replacement area of a segment that satisfies such conditions, the spare area management table 306 records addresses of sectors that can be used for replacement processing as shown in FIG. 19 in the form of the start and end of consecutive physical addresses. are doing.

Next, details of a method of selecting an alternative processing method will be described. FIG. 21 is a flowchart showing an example of the alternative process selection. Hereinafter, the operation of the alternative process selection will be described with reference to this flowchart. The substitute processing unit 302 receives a request for a substitute process from the control unit 303 and starts selecting a substitute process (S20).
1).

The substitution processing unit 302 refers to the substitution flag of the address conversion table 305, and confirms whether a replacement sector has already been assigned to the designated defective sector (S20).
2).

If an alternative sector has been allocated (ie, if the alternative flag is "1"), an alternative process in sector units is selected (S205).

If no substitute sector is assigned (ie, if the substitute flag is “0”), the control unit 30
For the logical addresses constituting the segment to which the defective sector specified by No. 3 belongs, the number of the replacement flags that are "1" is counted and compared with the threshold value.

When the number of the replacement flags "1" is less than the threshold value, the replacement processing in sector units is selected (S20).
5).

When the number of the replacement flags “1” is equal to or larger than the threshold value, the replacement process is selected in segment units (S
204).

The threshold value (Ns) is calculated as follows. The maximum access time for the segment permitted by the host device 300 is Tl, and when no defective sector has occurred, the time required for accessing the segment is T0, and the access time added by performing the alternative processing for one sector is Assuming that the maximum value is tmax, Ns = floor ((Tl-T0) / tmax) floor (x): Ns is determined as the largest integer less than or equal to x. Here, the maximum access time to the segment permitted by the host device 300 is set by the host device 300 when the disk device 101 is initialized.

The access speed of the disk section 304 differs depending on the storage position (access on the inner side of the disk is slow, and that on the outer side is fast). Thus, by setting the threshold based on the maximum access time allowed for each segment, the optimum threshold Ns for each segment is set.
, And as a result, the replacement storage block can be used efficiently.

Regarding the logical addresses constituting the segment, the number of which the replacement flag is “1” is counted, and the threshold Ns
By determining the alternative device as compared with the above, if the maximum access time (Tl) to the segment permitted by the host device 300 cannot be maintained in the sector unit replacement, a segment unit replacement process is performed. If the maximum access time can be kept, the alternative processing can be executed in sector units. As a result, the maximum access time to the segment is always maintained even after the replacement processing, and the number of replacement storage blocks to be used can be reduced.

(Embodiment 6) Embodiment 6 differs from FIG.
The configuration is the same as that of the disk device of the fifth embodiment. The only difference from the fifth embodiment is the method of selecting an alternative processing method and the method of selecting an alternative sector when performing the alternative processing in sector units. Here, only the method of selecting the alternative processing method and the method of selecting the alternative sector are described. Describe. FIG. 22 is a flowchart showing an example of the alternative process selection according to the sixth embodiment. Hereinafter, the operation of the alternative process selection will be described with reference to this flowchart. The substitute processing unit 302 receives a request for a substitute process from the control unit 303 and starts selecting a substitute process (S301).

Referring to the spare area management table, a substitute sector is provisionally assigned to a defective sector (S302).
A method of selecting an alternative sector will be described later.

The substitution processing unit 302 refers to the substitution flag of the address conversion table 305, and calculates the seek time and the rotation waiting time between the physical addresses before and after all the physical addresses for which the substitution processing in the sector unit is being executed. Asked,
The total value is calculated (S303). This sum is hereinafter referred to as a delay evaluation value. In the case of the example of the segment in FIG. 18, the seek time from a000 to j000 and the rotation waiting time,
The seek time from j000 to a002 and the rotation wait time are calculated in order, and the total sum is calculated. When the replacement flag is set for the sector at the end of the segment, the seek time and the rotation waiting time are obtained only in one direction in the segment. In addition, a value is similarly obtained for the temporarily assigned replacement sector.

By calculating the delay evaluation value in this way, it is possible to accurately grasp the variation in the access time due to the physical arrangement of each storage block used for replacement, and it is possible to obtain an unnecessary segment unit. The replacement process can be prevented, and as a result, the replacement sector can be used efficiently.

The delay evaluation value is compared with a threshold value (S30)
4).

When the delay evaluation value is less than the threshold value, the alternative processing for each sector is selected, and the temporarily assigned alternative sector is actually assigned (S307).

When the delay evaluation value is equal to or greater than the threshold value, the temporarily allocated alternative sector is returned to the spare area management table (S
305). Then, an alternative process for each segment is selected (S306).

The threshold value (Ts) is calculated as follows. Assuming that the maximum access time to the segment permitted by the host device 300 is Tl and the time required to access the segment when no defective sector is generated is T0, Ts is determined as Ts = Tl-T0. Here, the maximum access time to the segment permitted by the host device 300 is set by the host device 300 when the disk device 101 is initialized. The access speed of the disk unit 304 differs depending on the storage position (access on the inner side of the disk is slow, and access on the outer side is fast). In this manner, by setting the threshold value based on the maximum allowable access time of each segment, an optimum threshold value Ts can be obtained for each segment, and as a result, the alternative storage block can be used efficiently. It is possible to do.

Next, a method of selecting an alternative sector when temporarily assigning an alternative sector to a defective sector in (S302) will be described. The segment before the replacement sector assignment is performed is usually arranged at an adjacent physical address. However, the replacement sector cannot be located at a physical position adjacent to the physical address constituting the segment. Therefore, when a replacement sector is allocated, it is necessary to move the head to access the sector.
The disk unit 304 includes a rotating storage medium (disk) in which storage areas (sectors) are arranged on a circumference, and a head. One circumference on which a plurality of sectors are arranged is called a track, and a plurality of tracks exist concentrically on a disk. The head moves only on a straight line between the center of the disk and a point on the outer periphery of the disk (FIG. 23).
(O-O 'in FIG. 3), access to the sectors arranged on the circumference is performed until the target sector comes under the head by rotation of the disk. Therefore, the movement of the head is performed during the seek and the rotation wait. Seek refers to moving the head to the track where the target sector is located. The rotation waiting means waiting for a target sector to rotate to the position of the head among a plurality of sectors arranged on the track.

After the sector of the logical address (A0) in the segment is replaced by the replacement sector, when a segment access occurs, the head is moved after accessing the sector of the logical address (A0-1) and replaced. The sector of the logical address (A0) is accessed, and then the head is moved again to access the sector of the logical address (A0 + 1). Therefore, the substitute processing unit 302 calculates the seek time and the rotation waiting time of the movement from the sector of the logical address (A0-1) to the sector of the logical address (A0), and the logical address (A0 + 1) from the sector of the logical address (A0).
Then, a replacement sector that minimizes the sum of the seek time and the rotation waiting time for movement to the sector is selected. This method of selecting a replacement sector can reduce the access time when accessing a segment including a sector subjected to replacement processing. As a result, in many cases, it is possible to complete the access to the segment within a predetermined time even if the sector-by-sector replacement processing is performed, and the replacement sector can be used efficiently.

FIG. 2 shows a method of selecting this alternative sector.
3 will be described more specifically. FIG. 23 shows the structure of the disk of the disk unit 304. In the figure, the alphanumeric characters in parentheses in each sector represent logical addresses, and the other alphanumeric characters represent physical addresses. Now, consider the case where the sector of the physical address a012, which was the logical address (A012), has become a defective sector. As alternative sector candidates, there are nine sectors at physical addresses a027 to a035. As described above, when accessed as a segment after the substitution processing, the logical address (A0
The sector 11), the replaced sector of the logical address (A012), and the sector of the logical address (A013) are sequentially accessed while moving the head. Logical address (A
Considering the movement from the sector 011) to the replaced sector, first, it is necessary to perform a seek from the track to which the logical address (A011) belongs to the track where the replacement sector exists. Since the disk continues to rotate during the seek, the disk rotates the angle θ according to the seek time. In other words, when access to the track where the replacement sector exists is made after the access to the logical address (A011), access is possible from the sector moved by θ (the sector in the counterclockwise direction from OX) as shown in the figure.

Here, θ represents the number of rotations of the disk as R (rp
m), assuming that the seek time is S (sec), θ = 360 × 60 S / R is calculated. The rotation speed R of the disk is a value unique to the disk portion, and the seek time is a value that can be obtained if the source track and the destination track are known. Once it is known, θ can be obtained.

On the other hand, in the case where the logical address (A013) is accessed after the access to the substitute sector, the reduction of the rotation wait time after the seek is similarly considered.
It is desirable to select from sectors in the clockwise direction from OZ rotated by -θ from OY. If an attempt is made to access the sector at the physical address a013 after accessing the sector in the counterclockwise direction from OZ, for example, the sector at the physical address a035, the rotation latency is large because the physical block a013 has passed the head after the seek. Will be.

Therefore, in the example of FIG. 23, a sector suitable as a substitute sector for the logical address (A012) is OX
This is a sector that is more counterclockwise and simultaneously satisfies that it is a sector that is more clockwise than OZ. That is, the replacement sector may be selected from the physical addresses a032, a033, and a034. Whichever of the three physical addresses is selected, the time required for two head movements is the same.

In the fifth and sixth embodiments, the substitution processing unit 302, the spare area management unit 306, the address conversion unit 301, and the address conversion table 305 are provided in the disk device 101.
Side, a substitute processing unit 302, a spare area management table 306,
It is obvious that the same effect as that of the present embodiment can be obtained without using a special disk device by providing the address conversion unit 301 and the address conversion table 305 so that the host device 300 performs the substitution process. . Also, the substitution processing unit 302, the spare area management unit 306, the address conversion unit 30
1. The address conversion table 305 is stored in the disk device 101
And the host device 300, the sector unit replacement process is performed by the disk device, and the host device 300 executes the sector unit replacement process, thereby effectively utilizing the functions of a normal disk device, and It is obvious that the same effect as that of the present embodiment can be obtained without using a special disk device.

In the fifth and sixth embodiments, the information of the logical address constituting the segment (hereinafter referred to as the logical address information) in the logical address information in the access request from the host device 300
Segment configuration information) and detection of a defective sector is performed only when the host device 300 accesses the data. However, by adopting a method in which the host device 300 sets the segment configuration information when the disk device 101 is initialized,
Obviously, even when access requests other than the segment unit are mixedly generated to the disk device 101, an alternative process capable of keeping the maximum access time of the segment within a certain time can be executed.

In the fifth embodiment and the sixth embodiment, the storage device is a disk device. However, the present invention is similarly applicable to a storage device in which a storage area is constituted by storage blocks and which performs alternative processing. It is obvious that the same effects as in the present embodiment can be obtained even when applied to a tape device, a semiconductor memory device, and the like.

In the sixth embodiment, the delay evaluation value is obtained by summing the seek time and the rotation wait time between all the physical addresses in the segment where the substitution processing is being performed in the sector unit and the physical addresses before and after the physical address. However, if the difference between the physical addresses before and after the physical address of all the physical addresses in the segment where the substitution process is being performed in the segment is equal to or greater than a predetermined value, the value is set to “1”. Obviously, if the value is less than the value, the value is set to "0", and the total value is used as the delay evaluation value, so that the delay evaluation value can be easily obtained.
When the delay evaluation value described here is used, the threshold value (Ns) is calculated as follows.

Ns = floor ((Tl−T0) / tmax) floor (x): the largest integer less than or equal to x In the above equation, Tl is the maximum access time to the segment allowed by the host device 300, and T0 is 1 defective sector. When none occurs, the time required for accessing the segment, tmax is the maximum value of the access time added when the difference between the physical addresses becomes greater than or equal to a predetermined value once.

(Embodiment 7) Embodiment 7 of the present invention
Is a disk device having the same configuration as in the first embodiment. The operation will be described below with reference to FIG. 1 used in the first embodiment. Since the operation is almost the same as that of the first embodiment, different points will be described. First, (1) and (2) in FIG. 11 are the same as in the first embodiment. FIG. 11C shows an example in which (N−1) defects occur in one segment and (N−1) defective blocks are registered. FIG. 11D shows a case where an N-th defect occurs. At this time, the area between the respective defects (for example, the area between a11 and a22), the area between a10 and a12, and the area between aN1 and aN2 are compared. In this case, an area between two defective blocks aN1 and aN2 and the last block are selected. This area is combined into one and replaced with a continuous area of the alternative area. The method of selecting two defective blocks may be, for example, selecting the closest defective block. The method for registering each table is the same as that in the first embodiment, and a description thereof will be omitted.

FIG. 12 shows an example of a flowchart when the control unit 102 performs the substitution process.

As described above, the number of accesses to the replacement area in one continuous segment can be limited to N-1 times, and one segment can be read or written within a guaranteed fixed time. An excellent defective block replacement device that can be performed can be realized.

Embodiment 8 Embodiment 8 of the present invention
Is a disk device having the same configuration as in the first embodiment. The operation will be described below with reference to FIG. 1 used in the first embodiment. Since the operation is almost the same as that of the first embodiment, different points will be described. First, FIG. 13A is similar to the first embodiment. FIG. 13B shows an example in which the first defect has occurred in one segment and one defective block has been registered. a1
The number of blocks in the area between a10 and a12 is compared with the number of blocks in the area between a11 and a19, and the number of blocks in the area between a10 and a12 is small. In this case, a10 to a12
Are selected, and these are combined into one and replaced with a continuous area of the alternative area. FIG. 13C shows a case where a second defect occurs. In this case, a replacement area, a newly generated defective block, and blocks between them are selected, and these are combined into one and replaced with a continuous area of the replacement area. Also, when the third defect occurs, the same processing as when the second defect occurs can be performed, and thus the description is omitted. The registration method of each table is described in the first embodiment.
It is omitted because it is the same as.

FIG. 14 shows an example of a flowchart when the control unit 102 performs the substitution process.

As described above, the number of accesses to the replacement area in one continuous segment can be limited to one, and one segment can be read or written within a guaranteed fixed time. An excellent defective block replacement device that can be realized.

(Embodiment 9) Embodiment 9 of the present invention
Is a disk device having the same configuration as in the first embodiment. The operation will be described below with reference to FIG. 1 used in the first embodiment. Since the operation is almost the same as that of the first embodiment, different points will be described. First, (1) of FIG. 15 is a diagram after a defect occurs in which access is made N times or more in a segment, and a plurality of defective blocks are combined into one alternative area. FIG. 15 (2) shows a case where a defective block occurs in the replacement area. Alternative area
This is a case where defective blocks b5 to b6 occur in the substitute area from b1 to b2. In this case, the smallest alternative area is determined. In the present embodiment, the area between a11 and a21 and the number of area blocks between b1 and b6, the number of blocks between b1 and b2, and b5 and b2
And the number of blocks in the area between a and a32 to a42. An alternative process is performed by selecting a block having the smallest number of these blocks. This embodiment is an example in which b1 and b2 are the smallest. The result of performing the substitution process on the areas b3 to b4 is shown in (2). The method for registering each table is the same as that in the first embodiment, and a description thereof will be omitted.
Here, FIG. 16 shows an example of a flowchart when the control unit 102 performs the substitution process.

As described above, even if a defective block occurs in the replacement area, the number of accesses to the replacement area in one continuous segment can be limited to N, and one segment can be guaranteed to be a fixed number. An excellent defective block replacement device capable of reading or writing in time can be realized.

[0122]

As described above, according to the present invention, when a defect is newly detected, a storage block having a defect is registered as a defective block, and more than two storage blocks are registered in the continuous recording area to which the newly detected defect belongs. If there are already detected defects, two defective blocks included in the continuous storage area to which the newly detected defect belongs are selected, and the two defective blocks that are generated after the selection and the storage continuously arranged therebetween are selected. The access to a block is performed by accessing the segment arranged in the continuous area within a certain time by replacing the access to the recording block arranged continuously in the spare recording area and performing the continuous access. The remarkable effect that can be obtained is obtained.

In addition, by determining the replacement device using the information on the detected defective storage block and the information on the replacement storage block assigned to the storage block in the segment, the replacement in sector units can be performed. When the maximum access time for the segment permitted by the host device 300 cannot be maintained, the replacement processing is performed in segment units. When the maximum access time can be maintained even with the replacement in sector units, the replacement processing can be performed in sector units. As a result, a remarkable effect is obtained that the maximum access time to the segment is always maintained even after the replacement processing, and the number of replacement storage blocks to be used can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an embodiment of the present invention.

FIG. 2 is a schematic diagram for explaining Embodiment 1 of the present invention.

FIG. 3 is a diagram showing a specific example of a table for explaining the first embodiment of the present invention;

FIG. 4 is a flowchart illustrating Embodiment 1 of the present invention.

FIG. 5 is a schematic diagram for explaining Embodiment 2 of the present invention.

FIG. 6 is a diagram showing a specific example of a table for explaining the second embodiment of the present invention;

FIG. 7 is a flowchart illustrating a second embodiment of the present invention.

FIG. 8 is a schematic diagram for explaining Embodiment 3 of the present invention.

FIG. 9 is a flowchart illustrating a third embodiment of the present invention.

FIG. 10 is a flowchart illustrating a fourth embodiment of the present invention.

FIG. 11 is a schematic diagram for explaining Embodiment 7 of the present invention.

FIG. 12 is a flowchart illustrating a seventh embodiment of the present invention.

FIG. 13 is a schematic diagram for explaining Embodiment 8 of the present invention.

FIG. 14 is a flowchart illustrating Embodiment 8 of the present invention.

FIG. 15 is a schematic diagram for explaining Embodiment 9 of the present invention.

FIG. 16 is a flowchart illustrating Embodiment 9 of the present invention.

FIG. 17 is a configuration diagram of a disk device according to the fifth and sixth embodiments of the present invention.

FIG. 18 is a diagram showing the relationship between the configuration of an address management table and address conversion processing according to the fifth and sixth embodiments of the present invention.

FIG. 19 is a configuration diagram of a spare area management table according to the fifth and sixth embodiments of the present invention.

FIG. 20 is a flowchart showing an example of an alternative process according to the fifth and sixth embodiments of the present invention.

FIG. 21 is a flowchart showing an example of selection of an alternative processing method according to the fifth embodiment of the present invention.

FIG. 22 is a flowchart showing an example of selection of an alternative processing method according to the sixth embodiment of the present invention.

FIG. 23 is a structural diagram of a disk according to a sixth embodiment of the present invention.

FIG. 24 shows a conventional alternative device.

FIG. 25 is a diagram illustrating an example of a result of a replacement process performed by a conventional defective block replacement device.

[Explanation of symbols]

 Reference Signs List 101 disk device 102 control unit 103 address conversion unit 104 address conversion table 105 defect position management table 106 spare area management table 107 disk 201 memory 202 address generation 203 control unit 204 counter 300 host device 301 address conversion unit 302 substitution processing unit 303 control unit 304 disk unit 305 address conversion table 306 spare area management table

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Tsutomu Tanaka 1006 Kadoma Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (72) Inventor Masao Tamai 1006 Kadoma Kadoma, Kadoma City Osaka Pref. 72) Inventor Hirofumi Yokota 1006 Kazuma Kadoma, Kadoma-shi, Osaka Matsushita Electric Industrial Co., Ltd.F-term (reference) 5B065 BA01 CC03 EA15 5D044 BC01 CC04 DE03 DE38 DE62 DE64

Claims (17)

[Claims] 1. A storage device having a minimum recording unit of a recording device capable of random access and a plurality of the storage blocks. In an initial state, an access order of the storage blocks is continuous on the recording device, and It has a segment of a unit for performing access, and a spare recording area managed in units of an area in which the storage blocks are continuously arranged, and detects the storage block that has lost an initial function in access as a defective storage block. Detecting means; selecting means for selecting the storage block from the spare recording area as an alternative area when the defective storage block is detected; and storing the allocated alternative area with respect to access to the storage blocks constituting the segment. When accessing a block, access to the storage block in the alternative area is performed. A defective block replacement device having an access address control means for controlling an access position so as to perform one of the storage blocks, wherein the selection means deletes one storage block from the spare recording area based on an allocation state of a replacement area of the segment. A defective storage block replacement characterized by selecting at least two storage blocks from the preliminary recording area as replacement areas or replacing a plurality of storage blocks including the defective storage block. apparatus. 2. An alternative area selecting means as an alternative area for replacing the defective storage block in which the number of alternative areas in the segment is less than N and one storage block is detected from the spare recording area. If the number of the replacement areas is N in the segment, and the storage blocks are arranged in the order in which the read processing is executed in the segment, the detected defective storage blocks at both ends or Among the N sections surrounded by the adjacent replacement area including the replacement area or the detected defective storage block, the section having the smallest number of storage blocks existing in the section is detected, and the number of storage blocks having the smallest number is detected. An alternative area for replacing all the memory blocks existing in the smaller section is selected from an area continuously arranged in the preliminary recording area, and When the defective storage block is not included in the section having the smallest number of storage blocks, one storage block is selected from the preliminary recording area as a replacement area for replacing the detected defective storage block. Item 3. The defective memory block replacement device according to Item 1. 3. The replacement area selection means as a replacement area for replacing the defective storage block in which the number of replacement areas in the segment is less than N and one storage block is detected from the spare recording area. If the number of alternative areas in the segment is N, the storage blocks are arranged in the order in which the reading process is executed in the segment, and the first storage block and the last storage block of the segment are adjacent to each other. When considering an annular area, of the N + 1 sections surrounded by the adjacent defective area or the detected defective memory block including the detected defective memory block or the alternative area at both ends,
The section in which the number of storage blocks present in the section is the smallest is detected, and replacement areas for replacing all the storage blocks present in the section in which the number of storage blocks is the smallest are arranged consecutively in the preliminary recording area. Selecting from the area, when the detected defective storage block is not included in the section having the smallest number of storage blocks, substituting one defective storage block from the spare recording area for the detected defective storage block. 2. The defective memory block replacement device according to claim 1, wherein the device is selected as an area. 4. An alternative area selecting means, wherein the number of the alternative areas in the segment is less than N and one from the preliminary recording area
Storage blocks are selected as replacement areas for replacing the defective storage blocks, and if the number of replacement areas in the segment is N, the storage blocks are arranged in the order in which the reading process is executed in the segment. When the first storage block or the last storage block of the segment at both ends or the defective storage block or the replacement area detected including the head storage block or the last storage block or the replacement area of the adjacent segment including the detection area is detected. Among the N + 2 sections surrounded by defective storage blocks, the section having the smallest number of storage blocks existing in the section is detected, and all storage blocks existing in the section having the smallest number of storage blocks are replaced. The alternative area to be selected is selected from the areas continuously arranged in the preliminary recording area, When the issued defective storage block is not included in the section having the smallest number of storage blocks, one storage block is selected from the preliminary recording area as an alternative area for replacing the detected defective storage block. 2. The defect storage block replacement device according to claim 1, wherein: 5. The defective memory block replacement device according to claim 2, wherein the value of N is 2. 6. The method according to claim 1, wherein when the number of the replacement areas in the segment is one and the storage blocks are arranged in the order in which the reading process is executed in the segment, the replacement area and the replacement area are selected. A replacement area that replaces all the storage blocks included in the section surrounded by the defect storage block and the replacement area, including the defective storage block, is selected from the consecutively arranged areas of the spare recording area; If the number of replacement areas is 0 and the storage blocks are arranged in the order in which the reading process is executed in the segment, the first storage block or the last storage block of the segment and the detected defective storage block A section having a small number of storage blocks existing in the section is detected among the two sections enclosed, and the first storage block or the last storage block is detected. A substitute area selecting means for selecting a substitute area that replaces all the storage blocks existing in the detected section including the defective storage block from a continuous array of the spare storage areas. 2. The defect storage block replacement device according to claim 1. 7. When the occurrence of a defective storage block is detected in a segment, the alternative area selecting means includes information on the detected defective storage block and an alternative area assigned to a storage block in the segment. Using the information regarding the spare storage area, one storage block is selected from the spare recording area as a replacement area for replacing the defective storage block, or a replacement area for replacing the entire segment is stored in the storage block of the spare area. 2. The defective memory block replacement device according to claim 1, wherein the defective memory block replacement device is selected from continuously arranged regions. 8. The replacement area for replacing a segment is an area in which reading and writing of the entire replacement area can be executed within a predetermined time.
The defect storage block replacement device according to the above description. 9. The replacement area selection means sets the number of defects in the area to zero when no storage block for replacement is assigned to the storage block in the segment or when the entire segment is replaced by the replacement area. When a replacement storage block is assigned to a storage block in a replacement area in which a segment or an entire segment is replaced, the number of storage blocks to which the replacement storage block is assigned is determined by the number of defects in the area. When detecting the occurrence of a defective storage block in a segment, when the number of defects in the area is less than a predetermined threshold value or when the detected defective storage block is an alternative area consisting of one storage block, Selecting one storage block from the spare area as a replacement area for replacing only the detected defective storage block; When the number of internal defects is equal to or greater than a predetermined threshold value and the detected defective storage block is not a replacement area composed of one storage block, the replacement area for replacing the entire segment is replaced with the storage block of the spare area. 8. The defective memory block replacement device according to claim 7, wherein the defective memory block replacement device is selected from a region arranged in a row. 10. A substitute area selecting means, when an occurrence of a defective storage block is detected in a segment, temporarily selects one storage block of a spare area as a substitute area candidate for the defective storage block. A delay evaluation value, which is an evaluation value of a time required for access, is obtained using address information of each storage block in which data is actually recorded in a segment, and when the delay evaluation value is smaller than a predetermined threshold value. The alternative area candidate is selected as an alternative area that replaces only the detected defective storage block, and when the number of defects in the area is equal to or greater than a predetermined threshold, the area in which the storage blocks of the spare area are continuously arranged 8. The defect storage block replacement device according to claim 7, wherein a replacement area is selected. 11. The defective memory block replacement device according to claim 9, wherein the threshold value is set based on a permissible execution time of reading and writing of the entire storage area. 12. The defective storage block replacement device according to claim 1, wherein the storage block that is not a defective storage block among the replaced storage blocks is the storage block that is normally used. . 13. The defective storage block replacement apparatus according to claim 1, wherein the storage block that is not a defective storage block among the replaced storage blocks is registered in the spare recording area. 14. The storage device according to claim 1, wherein the area in which the storage blocks are continuously arranged is an area where a physical address or a physical position of the storage block is adjacent or close. The defect storage block replacement device according to the above description. 15. A method for selecting a replacement area for replacing a defective storage block in a storage system composed of a plurality of storage blocks, comprising the steps of: when occurrence of a defective storage block is detected, determining a logical address of the detected defective storage block; (A0)
Storage block of logical address (A0-1), storage block of logical address (A0), logical address (A0 + 1)
A method of selecting a storage block in a spare storage area, which has the shortest execution time when reading or writing the storage blocks sequentially, is selected as a replacement area for the defective storage block. 16. When a substitute area selecting means selects a substitute area for replacing a defective storage block from a spare area,
When the occurrence of a defective storage block is detected, and the detected logical address of the defective storage block is (A0),
Storage block of logical address (A0-1), storage block of logical address (A0), logical address (A0 + 1)
15. The storage block of the spare storage area, which has the shortest execution time when reading or writing the storage blocks sequentially, is selected as an alternative area for the defective storage block. 3. The defect storage block replacement device according to item 1. 17. The apparatus according to claim 1, wherein said alternative area selecting means is used for a magnetic disk.
7. The defective memory block replacement device according to any one of 6.