JPH02113467A - Disk control system - Google Patents

Abstract

PURPOSE:To eliminate the delay of response and to improve the throughput of a system by providing a disk controller with a memory where data of an alternative sector which cannot be assigned in the same track as a defect posi tion on a medium is stored. CONSTITUTION:When a power source is turned on or a disk driving device 13 is changed to a ready state, a magnetic disk controller 11 reads a correspondence table stored on a disk medium 14 and stores it in the memory of the controller 11. Simultaneously, the controller 11 reads the data of the alternative sector from the medium and stores it in an alternative sector data storage memory 12 of the controller 11. When a higher-order device 15 makes access to the medium, the controller 11 uses the read correspondence table and the data of the alternative sector which cannot be assigned in the same track as a defective sector read into the memory 12 to control the device 13 and responds to the device 15. Thus, the delay of response is eliminated to improve the throughput of the system.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to processing of a replacement sector corresponding to a defective sector in a disk device, and in particular, it is possible to improve the performance of access processing of the replacement sector. This paper relates to a disk control method.

[Conventional technology]

In a disk device, when a defective part of a medium is replaced with another good part, it is generally done on a sector by sector basis. This alternative sector (hereinafter referred to as replacement sector) does not reduce access processing performance, so
It is generally assigned to the same track or the same cylinder as the defective location. However, if it is impossible to allocate a replacement sector to the same track or cylinder because there are more than a certain number of defective sectors in the same track or cylinder, the conventional disk device For sectors that exceed the specified number, replacement sectors are allocated to other areas on the medium, or for tracks that have more than the specified number of defective sectors, this track is regarded as a defective track and replaced with another good track. It was to be allocated as a track. In addition, this conventional technology
The correspondence between defective sectors and replacement sectors is recorded on the medium and also stored in the disk control device. The address of the replacement area is checked by searching the correspondence table provided, and if a defective area is found, that area is accessed.

Incidentally, as a conventional technique related to this type of disk device, for example, a technique described in Japanese Patent Application Laid-Open No. 61-272821 is known.

[Problem to be solved by the invention]

In the above-mentioned conventional technology, in particular, when a replacement sector is not allocated to the same track or the same cylinder as a defective sector, or when a replacement track is not allocated to the same cylinder as a defective track, When an access instruction is issued to an area containing a defective sector, a mechanical seek operation is required to access the replacement sector or replacement track, causing a significant time delay in response to the host device and reducing throughput. There was a problem in that it decreased.

SUMMARY OF THE INVENTION An object of the present invention is to solve the problems of the prior art described above, and even when a replacement sector or a replacement track is assigned to a position that takes time to access, it can be accessed in the same amount of time as accessing a normal area. The object of the present invention is to provide a disk control method that enables response to a host device and does not reduce throughput.

[Means to solve the problem]

According to the present invention, the object is to include, in the disk controller,
This is achieved by providing a memory that stores data of a replacement sector or track that could not be allocated within the same track or cylinder as the defective location on the medium.

[Operation] When the power is turned on or the disk device shifts from the not-ready state to the ready state, the disk control device determines the correspondence between defective sectors written on the medium and replacement sectors, or between defective tracks and replacement tracks. At the same time, data of the replacement sectors or tracks that could not be allocated in the same track or the same cylinder as the defective part on the medium, i.e., data that takes time to access, is read into the memory in the self-control device. The data of the spare sector or track assigned to the position is read in advance and stored in the memory within the self-control device.

This allows the disk control device to access from a host device an area containing a defective sector where the replacement sector is not in the same track or the same cylinder, or an area containing a defective track where the replacement track is not in the same cylinder. When a Read instruction is received, the designated data can be read from the memory within the self-control device and transferred to the host device without actually accessing a replacement sector or track on the medium. Further, when similarly receiving a write instruction, the disk control device can return a response to the host device simply by updating the data in the memory.

Therefore, by using the present invention, it is possible to improve the throughput of the system.

〔Example〕

Hereinafter, one embodiment of the disk control method according to the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a magnetic disk system to which the present invention is applied, FIG. 2 is a diagram explaining the track format on the magnetic disk medium, and FIG. 3 is a diagram showing defective sectors and replacement sectors on the magnetic disk medium. FIG. 4 is a flowchart explaining the write processing operation, and FIG. 5 is a flowchart explaining the read processing operation. In FIGS. 1 and 2, 11 is a magnetic disk control device, 12 is a memory for storing alternate sector data, and 1
3 is a magnetic disk drive device; 14 is a magnetic disk medium;
15 is a host device, 16.17 is a backup battery power source, and 22.23 is a replacement sector.

In FIG. 1, a magnetic disk control device 11 controls a magnetic disk drive device 13 based on a command from a host computer or other host device 15, and controls a magnetic disk medium 1.
Access 4.

The magnetic disk control device 11 is equipped with a memory 12 for storing alternate sector data, and the magnetic disk control device 11 and the magnetic disk drive device 13 are each equipped with backup battery power sources 16 and 17, so that even after the power is cut off, , which can be operated for a certain period of time.

As shown in FIG. 2, each track on the magnetic disk medium 14 is divided into information recording/reproduction units called sectors. The sectors are classified into user sectors 21 in which data is recorded, and replacement sectors 22 which are used as a replacement if one of the user sectors 21 is defective. Normally, one or several replacement sectors 22 are provided on each track, and this is sufficient to relieve defective sectors on each track, but in rare cases, the number of replacement sectors 22 prepared is equal to the number of defective sectors. The number may exceed 22. In this case, the excess replacement sector 23 is allocated to a replacement sector track different from the track containing the defective sector prepared at a predetermined position on the magnetic disk medium 14.

The spare sector data storage memory 12 in the magnetic disk control device 11 holds the data of the spare sector when the spare sector is assigned to a track different from the track containing the defective sector.

The correspondence between the user sector 21 recognized as a defective sector and the replacement sector 22, 23 is determined when formatting the magnetic disk medium 14, for example, as a correspondence table 30 as shown in FIG. recorded and maintained.

The magnetic disk control device 11 is activated when the power is turned on or when
When the magnetic disk drive device 13 changes from the not-ready state to the ready state, it reads the above-mentioned correspondence table 30 recorded on the magnetic disk medium 14, stores it in the memory in the self-control device, and also identifies defective sectors. Regarding the corresponding replacement sector 23' that could not be allocated on the same track, the data is read from the magnetic disk medium 14 and stored in the replacement sector data storage memory 12 in the self-control key device 11.

In the above-mentioned state, if there is an access from the host device, the magnetic disk control device 11 will check the correspondence table 30 read into the magnetic disk control device 11 and the defective sector read into the replacement sector data storage memory 12. The data of the corresponding spare sector 23 that could not be allocated on the same track is used to control the magnetic disk drive device 13 and respond to the host device 15.

The operation of the magnetic disk control device 11 in response to access from the host device 15 will be described in detail below.

First, the processing when receiving the Wrij (3 instruction) from the host device will be explained with reference to the flow shown in FIG.

(1) First, the magnetic disk control device 11 receives a write instruction and write data from the host device, searches the correspondence table 30 stored in the internal memory, and writes the W
It is checked whether there is a defective sector within the target range of the rite process (steps 41 and 42).

(2) As a result of step 42, sectors with no defects are transferred to the user sector 21 on the magnetic disk medium 14;
For defective sectors, the data received from the host device 15 is written to the corresponding replacement sectors 22, 23 on the magnetic disk medium 14 (step 43).

(3) First, check whether there are any defective sectors existing within the write processing target range whose corresponding replacement sectors are assigned to a track different from the track where the defective sector is located, and Determine whether there is nothing (Step 4)
4.45).

(4) If it is determined in step 45 that there is a corresponding defective sector, the corresponding replacement sector data in the replacement sector data storage memory 12 is updated and the process ends (step 46).

(5) If it is determined in step 45 that there is no corresponding defective sector, the process is terminated without updating the corresponding replacement sector data in the replacement sector data storage memory 12.

In the process flow when the above-mentioned Write instruction is received, in step 43, even if the replacement sector corresponding to the defective sector is not allocated on the same track as the defective sector, other sectors are allocated to this replacement sector. However, for such a replacement sector, data is written only to the replacement sector data storage memory 12, and data is actually written onto the magnetic disk medium 14 at a later time. , it may be performed when the magnetic disk control device 11 and the magnetic disk drive device 13 have no more processes to execute. In this way, the magnetic disk control device 11 can avoid accessing the above-mentioned replacement sector (and send the Wr.
It is possible to report the completion of ite processing, so Wr
It is possible to improve the throughput of ite processing.

In this case, the backup battery 16. in the magnetic disk control device 11 and the magnetic disk drive device 13.
17 is unnecessary.

Furthermore, if the magnetic disk control device 11 and the magnetic disk drive device 13 are provided with backup battery power sources 16 and 17, the spare sector data stored in the spare sector data storage memory 12 can be transferred to the magnetic disk control device 11.
1 or all at once when the magnetic disk drive device 13 is powered off.

Further, the backup battery power source is provided only in the magnetic disk control device 11, and the backup battery power source is provided only in the magnetic disk control device 11.
1 may have a minimum capacity sufficient to hold the contents of the spare sector data storage memory 12 during power-off. In this case, even when the magnetic disk controller 11 is powered off, the contents of the spare sector data storage memory 12 can be retained, so the magnetic disk controller 11
When the power is restored, the data held in the spare sector data storage memory 12 can be written to the spare sector on the magnetic disk medium 14.

Next, processing when a Read instruction is received from a host device will be explained using the flow shown in FIG.

fl), [When the disk control device 11 first receives a Read instruction from the host device, it searches the correspondence table 30 stored in the internal memory and determines whether there is a defective sector within the target range of the Read process. (Step 51)
).

(2) As a result of step 51, data is read from the user sector 21 on the magnetic disk medium 14 for non-defective sectors, and even for defective sectors, corresponding replacement sectors are allocated on the same track. If so, data is read from that replacement sector 22 (step 52).

(3) Next, in the defective sector existing within the Read processing target range, the corresponding replacement sector is the same as the defective sector.
・Check whether there are any unallocated sectors on the rack and determine whether there are any corresponding defective sectors (
Steps 53.54).

(4) If it is determined in step 54 that there are no defective sectors, the read data is transferred to the host device 15.
and the process ends (step 55).

(5) If it is determined in step 54 that the corresponding defective sector exists, for the corresponding defective sector,
The corresponding spare sector data held in the spare sector data storage memory 12 in the magnetic disk controller 11 is read out without accessing the corresponding spare sector on the magnetic disk medium 14, and this data is stored in step 52. It is transferred together with the read data to the host device 15, and the process ends (step 56).

In the embodiment of the present invention described above, each track is provided with a replacement sector corresponding to a defective sector, and when the number of defective sectors exceeding the number of replacement sectors on each track occurs, the excess number of defective sectors is replaced with that sector. The present invention is applied to a method in which spare sectors corresponding to The present invention can be similarly applied to a method of allocating corresponding replacement sectors to other areas of the magnetic disk medium 14 for the excess number of defective sectors when more than 100 defective sectors occur in that cylinder. can.

In the embodiment of the present invention described above, when the magnetic disk control device 11 controls a plurality of magnetic disk drive devices 13, the alternate sector data storage memory 12 includes an area corresponding to each of the magnetic disk drive devices 13. It consists of

According to the embodiment of the present invention described above, no matter what area on the magnetic disk medium a replacement sector for a defective sector is allocated, the magnetic disk control device 11 causes the magnetic disk drive device 13 to [Effects of the Invention] As explained above, according to the present invention, the magnetic disk control device receives the defective sector from the host device in a short time. Rea of the area containing
When the d instruction is received, even if the replacement sector corresponding to the defective sector does not exist in the same track or the same cylinder as the defective sector, the host device does not have to perform the mechanical seek operation to the replacement sector again. Data can be transferred quickly and without delay.

In addition, when the magnetic disk control device receives a Wrii6 instruction for an area including a defective sector from a host device, if a replacement sector corresponding to the defective sector does not exist in the same track or cylinder as the defective sector, the magnetic disk Data can be written to the above replacement sector after a response is returned to the higher-level device, so there is no delay in responding to the higher-level device.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of a magnetic disk system to which the present invention is applied, FIG. 2 is a diagram explaining the track format on the magnetic disk medium, and FIG. 3 is a diagram showing defective sectors and replacement sectors on the magnetic disk medium. FIG. 4 is a flowchart explaining the write processing operation, and FIG. 5 is a flowchart explaining the read processing operation. 11...Magnetic disk control device, 12...
...Memory for storing alternate sector data, 13...
Magnetic disk drive device, 14...Magnetic disk medium, 15...Host device, 16.17...
・・Backup battery power supply, 22.23・・・・
...Replacement sector. Figure 2 Figure 3

Claims (1)

[Claims] 1. In a disk control method that allocates a replacement sector to another good location on the medium in response to a defective sector on the medium, a replacement sector may be placed on the same track or in the same cylinder as the defective sector. A memory for storing data of a replacement sector corresponding to the defective sector that could not be allocated is provided in the disk control device, and when a read instruction for an area including a defective sector is received from a higher-level device, the data of a replacement sector corresponding to the defective sector that could not be allocated is stored. If a replacement sector corresponding to the defective sector does not exist in the same track or in the same cylinder as the defective sector, the replacement sector is not accessed in the memory without accessing the replacement sector on the medium. A disk control method characterized by reading data at a corresponding location and transferring it to a host device. 2. In a disk control method that allocates a replacement sector to another good location on the medium in response to a defective sector on the medium, it is not possible to allocate a replacement sector to the same track or the same cylinder as the defective sector. A memory for storing data of a replacement sector corresponding to the defective sector is provided in the disk control device, and when a write instruction is received from a host device to an area including the defective sector, the replacement sector corresponding to the defective sector is stored. , if the defective sector does not exist in the same track or in the same cylinder, the replacement sector is written to the corresponding location in the memory without writing data to the replacement sector on the medium. A disk control method that is characterized by returning a response to the host device only by updating data. 3. Writing data to the spare sector on the medium where data has not been written to the spare sector is performed at a later time based on the data in the memory. Disk control method described. 4. By providing a backup battery power source in the disk control device and the disk drive, the data in the memory can be written to a spare sector in which data has not been written to a spare sector on the medium. 3. The disk control method according to claim 2, wherein the disk control method is performed when the disk control device is powered off or when the disk drive device is powered off. 5. By providing a backup battery power source that causes the disk control device to retain the contents of the memory, it is possible to prevent data from being written to a spare sector on the medium where data has not been written to the spare sector. 3. The disk control method according to claim 2, wherein the disk control method is performed based on the data after power restoration of the disk control device.