JP2001222866A - Reallotting method of logic address, data storage system and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data storage system in which even when a part of a storage region is made to be an unusable state the other usable storage region can be utilized again, in the case that the storage region of a whole system comprises storage regions which are physically divided into plural regions. SOLUTION: This system has such a function that normally, in the case that a physical storage regions is allotted to logically continuous addresses based on some regulation and controlled by a logical address of which the physical arrangement cannot be recognized from a high-order host, when a part of storage region physically divided is made to be an unusable state, logically continuous addresses are allotted again with a physically divided storage region unit.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for reassigning a logical address of a storage device of an information processing system and a data storage system, and more particularly to a case where a storage region of the entire system is composed of physically divided storage regions. The present invention relates to a logical address reassignment method for reusing another usable storage area even when a part of the storage area becomes unusable, a data storage system, and a recording medium.

[0002]

2. Description of the Related Art In a conventional data storage system of this type, when only a part of a storage area is unusable, the storage area is regarded as an unusable storage area and another storage area is used as an alternative storage area. It is generally performed to make the entire system usable by using it.

[0003] For example, when the entire physically divided storage area becomes unusable due to a failure or the like,
In situations where there is a large amount of unusable storage,
It is most desirable to replace an entire unusable physically separate storage area.

[0004]

However, in a magnetic disk device including a plurality of heads and media,
In the case of a system that is difficult to replace, such as when only some of the heads become unusable, using a method that replaces the entire unusable physically separated storage area,
Extremely difficult.

[0005] For this reason, when exchange is impossible, the entire system becomes unusable despite the fact that another physically separate storage area is available.

Also, when a method of replacing a defective sector in a storage area with a substitute sector prepared in advance is used, the number of substitute sectors is generally limited. When a bad sector exists, the number of replacement sectors for replacing all the bad sectors usually does not exist, so that there is a problem that the system cannot be recovered even by the method using the replacement sector.

SUMMARY OF THE INVENTION Accordingly, the present invention has been made in view of the above-mentioned problems, and has as its object to provide a partial storage area when the storage area of the entire system is physically divided into a plurality of storage areas. It is an object of the present invention to provide a method, a system, and a recording medium that allow another usable storage area to be reused even when the storage device becomes unusable. Other objects, features, advantages, and the like of the present invention will be immediately apparent to those skilled in the art from the descriptions of the following embodiments and the like.

[0008]

According to the present invention for achieving the above object, the storage area of the entire system is physically composed of a plurality of storage areas, and the physical storage area is usually based on a certain rule. Is assigned to a logically continuous address, and is managed by a logical address whose physical arrangement cannot be recognized from the upper-level host. In the case where the address becomes unusable, logically continuous addresses are re-assigned in physically separated storage area units.

Normally, recorded data is erased at the time of reassignment. However, in addition to the method of erasing data, reassignment is performed with data remaining.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In order to clarify the above and other objects, features and advantages of the present invention, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Referring to FIG. 1, there is shown a block diagram of a data storage system 2 according to a first embodiment of the present invention. The data storage system 2 has a control unit 21 for controlling the data storage unit 23, and a storage unit 22 for storing various information required when the control unit 21 controls the data storage unit 23.

The data storage means 23 stores data in the storage area 1 (23
1), storage area 2 (232),..., Storage area n (23)
n) (where n is an arbitrary integer of 2 or more)
It is physically divided into a plurality of storage areas. The data storage system 2 is controlled by the host 1.

Referring to FIG. 2, each storage area included in the data storage means 23, which is a component of the data storage system 2 of the present embodiment, has a storage area 1 of blocks 1 to m1, and a storage area 2 of blocks. The block 1 to block m2 and the storage area n are each composed of m1, m2,... Mn physical blocks, such as block 1 to block mn. Here, the physical block represents a minimum data recording unit in a storage area that can be controlled by the host 1 in the data storage system 2.

Referring to FIG. 4, there is shown a block diagram of a data storage system according to a second embodiment of the present invention. According to the configuration of the second embodiment of the present invention, the storage area, which is a component of the data storage unit 23, can be in an unusable state or an unusable state, as compared with the first embodiment. Detecting means 24 for detecting whether or not there is a sexual condition.

The operation of the first embodiment of the present invention will be described. The data storage system 2 according to the first embodiment of the present invention is controlled by the host 1. Data storage means 23 which is a component of the data storage system 2
Consists of a plurality of physically separated storage areas 1, storage area 2, storage area 3,..., Storage area n. Where n
Is an arbitrary number of 2 or more. Each storage area is composed of m1, m2, m3,... Mn physical blocks as shown in FIG.

The data storage system 2 has (m1 + m2
+ ... + mn) physical blocks.

Generally, in the normal use state, the host 1 cannot directly read and write data by designating this physical block.

The host 1 writes data into and reads data from the data storage system 2 by designating logically defined addresses in advance.

Here, the logically defined address means a logically continuous address (Logical Blcok Address, hereinafter abbreviated as “LBA”) based on a set rule. The above (m1 + m2 + ... +
mn) physical blocks. LB
The conversion from A to the physical block is performed by the control unit 21, and the conversion information is stored in the storage unit 22.

The assignment of a physical block to an LBA based on a predetermined rule is referred to herein as a “format”.

In the first embodiment of the present invention, one of the storage areas 1 (231) to n (23n), or a plurality of physically separated storage areas may be stored for some reason. When the storage area becomes unusable, a new LBA is re-allocated to a physical block (reformatting) with the storage area removed.

Also, in units of physically divided storage areas,
The control means 21 determines whether or not it has become unusable.

The operation of the control means 21 will be described with reference to the flowchart of FIG.

First, in response to a data read or data write request from the host 1, the control means 21 makes preparations for accessing the target LBA.

First, it is determined whether an error has occurred during data access preparation (step 302).

If there is no error, data read or data write requested by the host 1 is performed (step 303), and it is determined whether an error has occurred in the read / write processing (step 304). If there is no error, the process ends.

If an error occurs during LBA access from the host, such as during data access preparation or data read or data write, error information is stored in the storage means 22 (step 305).

Thereafter, it is determined whether or not there is a storage area satisfying a previously set unusable determination condition among the physically divided storage areas (step 30).
6).

If there is no condition that satisfies the unusable determination condition, the process is terminated.

If there is one that satisfies the unusable determination condition, the control means 21 reports to the host 1 that an unusable area has occurred (step 307).

Here, the host 1 determines whether or not a physically divided storage area determined as an unusable storage area is actually determined as an unusable area. However, this does not apply if the control unit 21 detects in advance that an unusable area has occurred and automatically sets the area as an unusable area.

In any case, if it is determined that the area is an unusable area, the upper-level host 1 reassigns the LBA at the timing determined.
This is performed by the control means 21.

Next, a second embodiment of the present invention will be described with reference to the drawings. As shown in FIG.
In the components of this embodiment, a detecting means 24 is further added to the configuration of the first embodiment. According to the second embodiment of the present invention, in addition to the operation of the first embodiment, it is possible to detect whether an unusable area has occurred in a storage area even when there is no data access. it can. At this time, data access is possible and it is a usable storage area, but it is detected whether there is a storage area that is likely to become inaccessible in the future, and before the error actually occurs, , Data can be saved and the area can be made an unusable area.

In the embodiment of the present invention, the control means 21 for controlling the data storage means 23 and the detection means 24 for detecting an abnormality in the storage area realize their functions and processes by program control of necessary control circuits. can do.

[0035]

EXAMPLES In order to explain the above-mentioned embodiment of the present invention in more detail, examples of the present invention will be described in detail.

As one embodiment of the present invention, FIG.
For example, a data storage device including a plurality of heads and a medium such as a magnetic disk device will be described. The host 1 is composed of a computer or the like that connects this data storage device.

The control means 21 is preferably a hard disk controller (hereinafter abbreviated as "HDC") and its peripheral circuits, and the storage means 22 is preferably readable and writable. More specifically, a read only memory (ROM) ), Random access memory (RA
M), a semiconductor memory such as a nonvolatile flash memory. Although there is a problem in terms of processing speed and reliability, various recording media such as a magnetic disk device can be used as the storage unit in terms of storage capacity and cost.

As the data storage means 23, a head disk assembly (Hard Disk Assembly; hereinafter, referred to as "HDA") in which a recording / reproducing head and a recording medium are combined is used.
And so on.

In this embodiment, one physically divided storage area corresponds to a combination of a pair of recording / reproducing heads and a medium.

In a data storage system in which a plurality of data storage devices typified by a disk array or the like are combined, one data storage device corresponds to one physically divided storage area, and 21 is
Corresponds to the disk array controller.

Hereinafter, a more specific case of the magnetic disk drive will be described. FIG. 9 schematically shows an example of the relationship between the recording / reproducing head and the corresponding cylinder, track, and sector on the medium surface.

Referring to FIG. 5, a general LBA format pattern when there are four recording / reproducing heads is shown as an example. Here, the numbered portions in the figure represent the tracks on the magnetic disk drive, and for each head, 0, 1, 2,..., N,.
Numbers are assigned in the order assigned to A.

In this example, as shown in FIG.
, The cylinder 0 of the head 1, the cylinder 0 of the head 2, the cylinder 0 of the head 3, the cylinder 1 of the head 0, the cylinder 1 of the head 1,... .

In each track, if there is no defective sector due to a medium defect or the like, it is general to assign LBAs to the sectors in order.

Since the track has a concentric structure, it is important to determine the position at which the sector number is assigned. Generally, a reference signal called an "index signal" is transmitted to a motor system circuit. This is used as a reference for assigning sector numbers.

More specifically, assuming that there are no bad sectors and that each cylinder contains k sectors, (head 0, cylinder 0, sector 0) → LBA: 0 (head 0 , Cylinder 0, sector 1) → LBA: 1 (head 0, cylinder 0, sector 2) → LBA: 2… (head 0, cylinder 0, sector k-1) → LBA: k-1 (head 1, cylinder 0) , Sector 0) → LBA: k (head 1, cylinder 0, sector 1) → LBA: k + 1… (head 0, cylinder 0, sector k-1) → LBA: 2k-1 (head 2, cylinder 0, Sector 0) → LBA: 2k (Head 2, cylinder 0, sector 1) → LBA: 2k + 1… (Head 3, cylinder 0, sector 0) → LBA: 3k (head 3, cylinder 0, sector 1) → LBA : 3k + 1… (Head 3, cylinder 0, sector k-1) → LBA: 4k-1 (head 0, cylinder 1, sector 0) → LBA: 4k (head 0, cylinder 1, sector 1) → LBA: 4k + 1… (Head 0, cylinder 1, sector k-1) → LBA 5k-1 (head 1, the cylinder 1, sector 0) → LBA: 5k (head 1, the cylinder 1, sector 1) → LBA: 5k + 1 ... a LBA assignment such a way that.

However, in general, the number of sectors is different for each cylinder, and the time difference at the time of head switching is set to a certain predetermined shift amount (skew). In many cases, the position of sector 0 is determined for each cylinder. Therefore, the position of sector 0 is not necessarily allocated to the same angle position on the concentric circle. Are shown in the figure.

When such a LBA assignment is used and a certain sector becomes unusable, the unusable LBA is assigned to a replacement sector provided in another place in advance. There is a method of re-assigning.

For example, when one of the heads becomes unusable due to a failure or the like, a large number of unusable sectors are generated.

As described in the background art, the number of replacement sectors is generally limited, so that when a large number of unusable sectors occur as in this case, The method cannot cope with this, and the entire device becomes unusable despite the availability of other heads.

Therefore, a case where the head 2 has failed will be described as an example. As shown in FIG.
A method of performing LBA allocation again (reformatting) may be considered.

According to this method, the apparatus can be reused by using only the heads other than the head 2.

At this time, it is desirable that the means for allocating the LBA, such as the HDC, determine whether the head has failed. By doing so, the head failure can be determined by the operation shown in the flowchart of FIG.

In this embodiment, the operation of the flowchart of FIG. 3 will be described. The data access preparation error (step 302) is, for example, a head seek error or other processing required before data access, and And errors depending on the physical position of the data to be accessed, such as a head failure.

The error information (step 305) is
The head number, cylinder number, sector number, etc. in which an error has occurred correspond to this.

Unusable condition (step 306)
For example, when the total number of errors in data read / written by one head having recorded error information reaches an arbitrary, preset number, and If a track occurs that is an error, or if such a track is any pre-defined,
When the number has been reached, there are others.

In the determination as to whether or not the unusable determination condition is satisfied, if the condition is satisfied, the host 1 such as a computer connected to the magnetic disk drive is used.
Is notified of the occurrence of an unusable area (step 30).
7).

Thereafter, whether the upper host 1 such as a computer connected to the data storage device performs the above-mentioned LBA reassignment (reformat),
Judgment is made on the timing of performing reassignment (reformatting).

If it is set in advance that the reformatting is automatically performed when an unusable area occurs, the reformatting may be performed without the judgment of the host 1.

Although not shown in the flowchart of FIG. 3, it is also possible to perform an operation of confirming whether or not the apparatus is unusable after satisfying the unusable determination condition.

This checking operation includes, for example, performing data reading at an arbitrary number of places and, if all data are errors, setting a head failure.

In any case, if it is determined that the area is an unusable area, the reassignment of the LBA is performed at the timing determined by the host 1.
C and the like, and the assignment information is stored in the storage means 22.

As another example, when the time required for one cylinder seek is shorter than the time required for head switching, a format pattern as shown in FIG. 7 may be used.

In this pattern, LBAs are allocated in the order of four tracks in the direction in which the cylinder number on the same head increases, and then, from the track of the head next to the same cylinder as the last allocated track, the cylinder number is This is a format pattern in which four tracks are allocated in the direction in which is smaller.

Also in this case, by the same operation as in FIG. 5, for example, when the head 2 becomes unusable, LBA reassignment (reformat) as shown in FIG.
, The remaining usable heads can be used.

Next, a second embodiment of the present invention will be described. The configuration of the second embodiment of the present invention is the same as that of the second embodiment described with reference to FIG. The detection by the detecting unit 24 in FIG. 4 means, for example, in a data storage device including a plurality of heads and a medium such as a magnetic disk device, a physical position of a data storage area such as a head or medium abnormality. It refers to detecting abnormalities in related means.

More specifically, for example, in the case of a magnetic disk drive in which a recording / reproducing head flies above a medium, a means for detecting or estimating a change in a head flying height which is an interval between the head and the medium is used. Therefore, if the head flying height is out of the range of the head flying height set in the normal state, the abnormality of the head flying height is reported to the control means 21 in FIG. As in the case where the error occurs, it is possible to perform reformatting so that the data storage area corresponding to the abnormal head is not used.

At this time, among the data to be read by the problematic head, the operation of saving the still readable data to the recording area of another head or saving to another magnetic disk device is performed. It is also possible.

In order to detect another abnormality, a certain data is reproduced for each storage area corresponding to each head, or an error rate which is the number of error bits with respect to the total number of read bits when recording and reproducing is measured. A method of determining whether or not there is an abnormality or a possibility of becoming abnormal in the near future according to the value of the error rate; or In a bad state, a method of performing a load test such as measuring an error rate or the like and making a determination can be used.

As described above, in both the first embodiment and the second embodiment, if the data recorded on the data storage system can be saved on another data storage system at the time of reformatting, , It is desirable to perform reformatting including a function of completely erasing data. This is because the lack of a part of the data causes the continuity of the data to be lost, resulting in incomplete data.

However, when a large amount of data such as image data is recorded, the remaining data may be reusable even if part of the data is lost.

In the case where such an advantage is obtained, it is possible to read out the data in an incomplete form with some data missing by reformatting without erasing the data. Become.

For example, in the case shown in FIG. 5, when the head 2 fails, the cylinders 0, 1 and 2 shown in FIG. 5 are re-formatted as shown in FIG.
3, 4, 5, 7, 8, 9, 11, 12, 13, 15, ...
It is possible to provide a possibility that a series of data except for the data recorded in the head 2 can be read.

Further, when a lump of data exists only on a head which has not failed, complete data can be reproduced.

Further, all sectors on each head are represented by L
Obviously, in the case of a format pattern in which a sector on the next head is allocated after the allocation to the BA, more data can be completely reproduced.

In each of the above embodiments, the operation of the control means 21 and the detection means 24 can be realized by a program executed by a computer that controls the data storage system 2. In this case, a recording medium (semiconductor memory, FD, MT, CD-
ROM), from a predetermined reading device as needed,
By reading the program into the data storage system 2 and loading the program into, for example, the main memory of the computer of the data storage system 2 and executing the program, the control unit 21 and the detection unit 24 in the data storage system according to the present invention.
Function can be realized.

The drawings described above are only for illustrating the present invention, not for limiting the scope of the present invention, and the present invention is not limited to each of the above embodiments. Principles of the invention defined in the claims
Needless to say, various changes and modifications that can be made by those skilled in the art are included within a range according to the spirit.

[0078]

As described above, according to the present invention,
The storage area of the entire system consists of physically divided storage areas, and usually, based on a certain rule,
When the physical storage area is assigned to logically continuous addresses and is managed by a logical address that cannot be recognized from the upper level host, the physical When a physically separated storage area becomes unusable, a function of reassigning logically continuous addresses again in units of physically separated storage areas can be used for other uses. This has the effect of making possible storage areas reusable.

Further, according to the present invention, by detecting an abnormality in each of a plurality of physically separated areas, data is evacuated before it becomes unusable, the area is made unusable, and data security is reduced. Is guaranteed.

Further, according to the present invention, it is possible to provide a function and a data storage system which can use data other than a lost part in the case of reusable data even if recorded data is partially lost. Can be done.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first exemplary embodiment of the present invention.

FIG. 2 is an explanatory diagram for explaining a storage area according to the first embodiment of the present invention.

FIG. 3 is a flowchart illustrating an example of an operation according to the first exemplary embodiment of the present invention.

FIG. 4 is a block diagram showing a configuration of a second exemplary embodiment of the present invention.

FIG. 5 is an explanatory diagram illustrating an example of a format pattern according to an embodiment of the present invention.

FIG. 6 is an explanatory diagram showing an example of reformatting in one embodiment of the present invention.

FIG. 7 is an explanatory diagram showing an example of a format pattern in one embodiment of the present invention.

FIG. 8 is an explanatory diagram showing an example of reformatting in one embodiment of the present invention.

FIG. 9 shows a head, a cylinder, and a magnetic disk drive.
FIG. 3 is an explanatory diagram illustrating an example of a relationship between a track and a sector.

[Explanation of symbols]

 Reference Signs List 1 upper host 2 data storage system 21 control means 22 storage means 23 data storage means 24 detection means 231 storage area 1 232 storage area 2 23n storage area n

Claims (12)

[Claims] 1. A storage system according to claim 1, wherein all storage areas are physically divided into a plurality of storage areas, and based on a predetermined rule.
A physical storage area is allocated to logically continuous addresses, and from the upper host, in the data storage method managed by the logical address, some physically separated storage areas cannot be used. A logical address reassignment method characterized by re-assigning logically continuous addresses again in units of physically divided storage areas when a state is reached. 2. The logical address reassignment method according to claim 1, wherein the reassignment of the logical address is also performed when it is determined that there is a possibility that a physically divided storage area becomes unusable. And a logical address reassignment method. 3. The logical address reassignment method according to claim 1, wherein even if recorded data is partially lost, reusable data is physically separated as unusable. A logical address reassignment method characterized by reusability by joining together in a form excluding a part recorded in a storage area. 4. The logical address reassignment method according to claim 2, wherein the data recorded on the physically separated storage area determined to be unusable may be replaced with another physical address. A logical address reassignment method that saves data on a separate storage area. 5. A storage system according to claim 1, wherein all storage areas are physically divided into a plurality of storage areas, and based on a predetermined rule.
Physical storage areas are assigned to logically consecutive addresses, and some physically separated storage areas cannot be used in a data storage system that is managed by the logical address from the host. A data storage system comprising means for re-assigning a logically continuous address again in units of physically divided storage areas when the state becomes a state. 6. The data storage system according to claim 5, wherein said logical address is reassigned even when it is determined that a physically divided storage area may become unusable. A data storage system, comprising: 7. The data storage system according to claim 5, wherein even if recorded data is partially lost, reusable data is stored in a physically separated storage area which is made unusable. A data storage system, characterized in that the data storage system comprises means for reusable by joining together in a form excluding portions recorded in the data storage. 8. The data storage system according to claim 5, wherein the data recorded on the physically divided storage area determined to be unusable is stored in another physically separated storage area. A data storage system, comprising: means for saving on a divided storage area. 9. A data storage system which is controlled by a host and reads and writes data from a data storage in response to a data write and read request from the host, wherein the data storage means comprises a plurality of physical blocks. A control unit that controls the data storage unit; and a storage unit that stores various information required when the control unit controls the data storage unit. An upper-level host, for the data storage system,
Data writing and reading are performed by designating a logically defined address. The storage unit stores a conversion rule from a logically continuous address to a physical block. In the case where it is determined that one of the plurality of storage areas or a plurality of physically separated storage areas has become unusable, the state in which the unusable storage area is removed A new logically continuous address being reassigned to a physical block. 10. A storage device comprising: a detection unit for detecting whether an unusable area has occurred in the storage area even when there is no data access to the data storage unit. Detects whether there is a storage area that is available but is likely to become inaccessible in the future, saves data before an error actually occurs, and makes that area unusable 10. The data storage system according to claim 9, wherein the data storage system is configured to be controlled to be an area. 11. A data storage system controlled by an upper host and reading and writing data from a data storage unit in response to a data write and read request from the upper host, wherein the data storage unit includes a plurality of physical blocks. A control unit that controls the data storage unit; and a storage unit that stores various information required when the control unit controls the data storage unit. An upper-level host, for the data storage system,
Data writing and reading are performed by designating a logically defined address, and the storage means stores a conversion rule from a logically continuous address to a physical block. In the case where it is determined that one of the plurality of storage areas or a plurality of physically separated storage areas has become unusable, the state in which the unusable storage area is removed A recording medium on which a process for newly reassigning logically continuous addresses to physical blocks is executed, and a program for causing a computer constituting the data storage system to execute the process is recorded. 12. A recording medium according to claim 11, wherein a process for detecting whether an unusable area has occurred in said storage area even when there is no data access to said data storage means. Data access is possible and it is a usable storage area, but it is detected whether there is a storage area that is likely to become inaccessible in the future, and data is evacuated before an error actually occurs. And a process for controlling the area to be an unusable area; and a recording medium storing a program for causing the computer constituting the data storage system to execute the respective processes.