JPH04266117A - Storage device - Google Patents

Abstract

PURPOSE:To reduce the labor and time and to prevent the loss of memory by estimating an area having the high possibility of the occurrence of a defect in a storage device based on number of times of the retrial and replacing automatically the estimated area with a substitute area. CONSTITUTION:A control table 5 to control a defective area and number of times of access retrial storage area 7 are provided. Then the number of times of retrial at every input/output operation and the number of times of total trial are controlled at storage area. An area having the high possibility of the occurrence of a defect is estimated when the number of times of retrial of the relevant area exceeds a set upper limit level. Then the estimated area is assigned to a substitute storage area. Thus the occurrence of defects can be prevented.

Description

[Detailed description of the invention]

0001

TECHNICAL FIELD The present invention relates to a storage device, and in particular,
For example, it relates to randomly accessible storage devices such as magnetic disk devices and magneto-optical disk devices.

0002

2. Description of the Related Art Conventional randomly accessible storage devices have a management table 5 for managing defective storage areas generated during the manufacturing process, as shown in FIG. has an alternative storage area 6 of
When a defective storage area occurs in the normal data storage area 7, this is used to replace the defective area by allocating it as a new storage area.

0003

[Problems to be Solved by the Invention] However, in the above conventional example, if a defect occurs while the storage device is in use, a person must intervene and register an alternative area in the device to clean up the defective storage area. did not become. Furthermore, if a failure occurs when attempting to read information from an area where information has already been stored, the information in that area will be lost.

The present invention has been made in view of the above conventional example, and provides a storage device that can predict the occurrence of a defective storage area, prevent input/output operation failures, and automatically replace a defective storage area. The purpose is to

[0005]

Means for Solving the Problems In order to achieve the above object, the storage device of the present invention has the following configuration. In other words, it is a random access type storage device that includes a storage area for storing information and an alternative storage area for responding to failures in access to the storage area, and the input/output operation is performed every time an input/output operation to the storage area occurs. a first checking means for checking whether the output operation is normal or defective; a second checking means for checking the total number of times the input/output operation is defective detected by the first checking means from a predetermined time; an alternative storage area allocating means for allocating an alternative storage area as a new storage area, and each time a failure in the input/output operation occurs, based on the check results of the first checking means and the second checking means, and a control means for controlling the input/output operation to be retried or for the alternative storage area allocating means to allocate an alternative storage area.

[0006]

[Operation] With the above configuration, the present invention can perform the following operations based on the check results of the first checking means and the second checking means
Control is performed to retry the input/output operation or to allocate an alternative storage area to the part where the input/output operation failure has occurred.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a block diagram showing the configuration of an information storage device that is a typical embodiment of the present invention. In FIG. 1, 1 is a hard disk serving as a medium for storing information, 2 is a central control unit (CPU), 3 is a ROM, and 4 is a RAM.
It is.

FIG. 2 is a diagram showing the concept of storage area state management of the hard disk 1 used by the information storage device of this embodiment. In FIG. 2, 5 is a management table for managing defective areas, 6 is an alternative storage area, and 7 is a normal data storage area. Furthermore, 8 is a storage table for the number of retries to access the hard disk 1, and this is used to predict the occurrence of a failure. Note that the management table 5 and the access retry count storage table 8 are placed at predetermined locations in the RAM 4 of the information storage device or the hard disk 1. According to Figure 2,
The data storage area 7 is divided into small storage areas serving as management units, and management information for defective areas among these is stored in the management table 5. This is an example in which the small storage areas of management units A-3 and B-1 are managed as defective storage areas, and a part of the alternative storage area 6 is allocated as an alternative area. The number of areas in the management table 5 is the same as the number of alternative areas 6 (that is, one-to-one correspondence). Further, the access retry count storage table 8 stores the access retry count for each management unit. Figure 2 shows A-4
In this example, there are a total of i access retries for C-3 and j access retries for C-3.

Next, the flowcharts shown in FIGS. 3 and 4 describe the process of predicting a defective storage area and allocating an alternative area during a write operation and a read operation using an information storage device that performs such a configuration and storage area state management. will be explained with reference to each. In addition, in FIGS. 3 and 4, re
try represents the number of retries for each input/output operation, RETRY represents the total number of retries, rmax represents the upper limit of the total number of retries, and Rmax represents the upper limit of the total number of retries. Here, it is assumed that at the beginning of processing, retry is set to "0", RETRY is set to the total number of retries up to the previous access operation, and rmax and Rmax are each set to an appropriate upper limit value.

First, the processing at the time of writing will be explained. First, in step S1, if there is a write operation, it is checked in step S2 whether or not there was an error in the write operation. Here, if no error occurs, the process ends, and if an error is detected, the process proceeds to step S3. In step S3, the value of retry is set to "1". Next, in step S4, the value of RETRY is updated (
+1). In step S5, retry and RETR
A check is performed by comparing Y with rmax and Rmax, respectively. Here, if either retry or RETRY exceeds the condition value, a defect has occurred (ret
ry≧rmax) or that there is a high possibility that a defect will occur (RETRY≧Rmax), the process proceeds to step S6, performs storage area replacement processing, and then returns to step S6.
At step 7, a write operation is performed at the newly substituted location. Subsequently, in steps S8 to S9, retry and RET
The value of RY is reset to "0", and a write operation to the alternative area is checked in step S10. here,
If there is no error, the process ends, and if there is an error, the process returns to step S3 to perform a retry process.

On the other hand, if both retry and RETRY are less than the upper limit, the process proceeds to step S11, and the write operation is retried. In step S12,
The value of retry is updated (+1), and then step S
At step 13, the retry result of the write operation is checked. Here, if there is an error, the process returns to step S4 and continues to retry the write operation; if there is no error, the process proceeds to step S14, where the value of RETRY is updated (+1), and finally, in step S15, the retry The value of is reset to "0" and the process ends.

As described above, in the write operation processing, R
By setting max>rmax, it is possible to avoid locations where errors occur frequently and are considered to be potential defects. Furthermore, even if a defect occurs, a replacement can be automatically performed.

Next, the processing at the time of reading will be explained. Note that processes common to the write process are given the same step numbers and their explanations are omitted.

First, following the read operation in step S21, the processes in steps S2 to S4 are performed, and then in step S25, retry is compared with rmax and checked. Here, if retry≧rmax, the process proceeds to step S26, and it is determined whether or not to execute alternative processing. If alternative processing is not performed, the read operation processing is terminated as an access error. On the other hand, if retry<rmax, the process proceeds to step S27 and retries the read operation. Next, in steps S12 and S13, the retry value is updated and the read operation is checked, and then the process proceeds to step S28, where the RE
Check by comparing TRY with Rmax. here,
If RETRY<Rmax, the read operation ends; however, if RETRY≧Rmax, the corresponding area currently being accessed is considered to be potentially defective, and the process proceeds to step S29, where the information in the corresponding area is replaced. Executes the process of moving to the area. In step S29, storage area replacement processing is performed, and in step S30, writing processing to the replacement area is performed.

The processing for the read operation described above is basically the same as the processing for the write operation, but the fact is that the occurrence of defects during the read operation is essentially unavoidable.
Writing and processing are different.

Therefore, according to this embodiment, it is possible to check the number of retries and the total number of retries every time an access operation occurs, and to allocate an alternative area based on the check results. Now, in this embodiment, the case where the access retry number storage table 8 is placed in one place has been explained.
The present invention is not limited to this. For example, Figure 5
As shown in FIG. 2, each storage area of the hard disk 1 may be provided with an area for storing information corresponding to the access retry number storage table 8. As a result, the process of obtaining the number of retries can be performed simultaneously with writing and reading, and the effort of providing a separate area and reading from there can be omitted. Furthermore, as shown in FIG. 6, it is also possible to store the number of retries only for areas where retries have been made, and not to record the number of retries for areas where no retries have been made. This makes it possible to save the area for recording the number of retries for a storage device with a low error rate.

Furthermore, in this embodiment, the total number of retries was checked every time an access operation occurred.
It is also possible to discover potential errors with a higher probability by reading and accessing all areas within the storage device at the time of startup of a system including the storage device or periodically.

The present invention may be applied to a system consisting of a plurality of devices, or may be applied to a device consisting of one device, or may be achieved by supplying a program to the system or device. Needless to say, this method can be applied in cases where

[0020]

Effects of the Invention As explained above, according to the present invention, by predicting in advance an area in a storage device where a failure is likely to occur and allocating it to an alternative storage area, input/output This has the effect of preventing malfunctions from occurring. Furthermore, even when an input/output malfunction occurs, by automatically replacing the area, it is possible to save the effort of manually performing the replacement.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an information storage device that is a typical embodiment of the present invention.

FIG. 2 is a diagram showing the concept of storage area state management.

FIG. 3 is a flowchart of defective area detection and defective area replacement processing during a write operation.

FIG. 4 is a flowchart of defective area detection and defective area replacement processing during a read operation.

FIG. 5 is a diagram illustrating the concept of storing storage area status management information in the same area as normal information.

FIG. 6 is a diagram illustrating the concept of reducing the storage area of storage area status management information.

FIG. 7 is a conceptual diagram of storage area state management according to a conventional example.

[Explanation of symbols]

1 Hard disk 2 CPU 3 ROM 4 RAM 5 Bad area management table 6 Alternative storage area 7. Normal data storage area

Claims (1)

[Claims] 1. A random access type storage device comprising a storage area for storing information and an alternative storage area for responding to failures in access to the storage area, wherein the random access storage device includes a storage area for storing information and an alternative storage area for responding to failures in access to the storage area, wherein a first checking means for checking whether the input/output operation is normal or defective; and a second checking means for checking the total number of times the input/output operation is defective detected by the first checking means from a predetermined time. and alternative storage area allocating means for allocating the alternative storage area as a new storage area; 1. A storage device comprising: control means for causing the input/output operation to be retried or causing the alternative storage area allocating means to allocate an alternative storage area based on the input/output operation.