JPS63288470A - Disk controller - Google Patents

Abstract

PURPOSE:To quickly perform alternative processing by providing a semiconductor memory and reading out block alternative information or sector alternative information from a magnetic disk device to store this information in the semiconductor memory and referring to the stored alternative information. CONSTITUTION:A disk controller 12 incorporates a semiconductor memory 12c, and a magnetic disk device 13 is so set at the time of formatting that data is written and read in block units. The disk controller 12 reads out block alternative information from the magnetic disk device 13 and a stores it in the semiconductor memory 12c. When a command to indicate data input/output is received from a host computer 11 stored block alternative information is referred to discriminate whether the block designated by the command is includ ed in block alternative information or not, and data is inputted to or outputted from an alternative block indicated by block alternative information if it is included there.

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to a disk control device and is intended to improve the efficiency of reading and writing information.

<Prior Art> FIG. 3 is a block diagram showing an example of an electronic computer system. In the figure, 1 is a host computer, 2 is a disk control device, and 3 is a magnetic disk device. When the host computer 1 outputs a command for reading or writing, the disk control device 2 outputs a command to operate the magnetic disk device 3. When the predetermined operation (bed seek, etc.) of the magnetic disk device 3 is completed, the status indicating that read or write preparation has been completed changes to
Output from the magnetic disk device 3 and sent to the disk controller 2
The data is input to the host computer 1 via the host computer 1. When such control is completed, the host computer 1 reads or writes data to the magnetic disk device 3 via the disk control device 2.

As shown in FIG. 4, the magnetic disk device 3 includes a plurality of disks D and a plurality of heads H. The plurality of disks D are fixed to the rotating shaft 4 and rotate as one,
A plurality of heads H are installed in a seek mechanism 5 and move together along the radial direction of the disk.

Therefore, as shown in FIG. 5, a plurality of concentric tracks T, which are information recording portions, are formed on each disk D, and the formation positions of the tracks T formed on each disk D are located at the rotation axis 4. The distances III) from the center of the upper and lower disks D are the same. Assuming a surface that connects the disks D with the same track formation positions vertically, it becomes a cylindrical surface, which is called a cylinder. In other words, you can specify a single track by determining the cylinder number and head number. Furthermore, each track T is divided into a plurality of areas called sectors, and the minimum unit of data storage is assigned to each sector for input/output. Note that the operation of writing a unique address to each sector is called formatting.

Data is written and read in units of sectors or blocks, depending on the system configuration. A block is one or more consecutive sectors on the same track. By performing input/output processing in units of blocks, it is possible to take advantage of the magnetic disk drive's advantage of efficient continuous data input/output to consecutive sectors on the same track.

By the way, due to variations during manufacturing, there are defective areas on the magnetic surface of the disk where data cannot be stored correctly even if written. This defective part cannot be used as a storage part. For this reason, the disk device 3 does not guarantee normal data input/output for all sectors, but assumes that some defective sectors will occur.

Here, conventional means for dealing with the above-mentioned bad sectors will be explained. FIG. 6 shows the state of the storage area of the magnetic disk device 3. As shown in the figure, Not~No (n −
The cylinder 2) is used as an input/output canister which is a normal storage section, and the two cylinders No. (n-1) and No. Sector replacement information is written in sectors with no defects among the NoO cylinders. Sector replacement information is information that indicates the correspondence relationship between defective sectors in the turnout caniglia and replacement sectors in the replacement area.
As shown in the figure, a defective sector number and a substitute sector number are stored as a memory table in which a pair constitutes a substitute unit.

When the disk control device 2 receives a command to instruct data input/output from the host computer 1, it first reads sector substitution information from the magnetic disk device 3 and checks whether the sector specified by the command is included in the sector substitution information. Determine. If the sector specified by the command is not included in the sector replacement information, data is input/output to the specified sector. When the sector specified by the command is included in the sector alternative information, data is input/output to the alternative sector indicated by the sector alternative information.

<Problems to be Solved by the Invention> In the conventional means for dealing with bad sectors, since sector replacement information is stored in the magnetic disk device 3, the disk control device 2 performs sector replacement to replace the bad sector with a replacement sector. It is necessary to refer to the control information area of the magnetic disk device 3 every time processing is performed. As a result, the load on the disk control device 2 increases and input/output efficiency decreases.

Also, when performing input/output in units of blocks, 1
Since a defective sector included in a block is replaced with an alternative sector, the normal sector in this block and the alternative sector become separated, making it impossible to perform continuous sector input/output.

As a result, the benefits of using a block as the input/output unit are reduced.

In view of the above-mentioned prior art, the present invention provides a disk control device that can quickly perform alternative processing.

Means for Solving the Problems> The present invention for solving the above problems includes a semiconductor memory provided in a disk control device, reading block replacement information or sector replacement information from a magnetic disk device and storing it in the semiconductor memory, It is characterized by performing alternative processing by referring to stored alternative information.

<Example> FIG. 1 shows an example of the present invention. In the figure, 11 is a host computer, 12 is a disk control device, and 13 is a magnetic disk device.

In addition to the data buffer 12a and processor 12b, the disk control device fi12 also includes a semiconductor memory 12c.
is built-in. In this example, the magnetic disk device 13
is set during formatting so that data can be written and read in blocks. Block replacement information for replacing a defective block with a replacement block is written in the control information area of the magnetic disk device 13. This block replacement information is stored as a memory table in which a defective block number and a replacement block number form a set to form a replacement unit, as shown in FIG.

In this embodiment, when the computer system starts operating, the disk control device 12 first reads block replacement information from the magnetic disk device 13 and stores it in the semiconductor memory 12c.

When the processor 12b of the disk control device 12 receives a command instructing data input/output from the host computer 11, by referring to the block replacement information stored in the semiconductor memory 12b, the block specified by the command is replaced with the block replacement information. Determine whether it is included in If the block specified by the command is not included in the block replacement information, data is input/output to the specified block. When the block specified by the command is included in the block replacement information, data is input/output to the replacement block indicated by the block replacement information. Note that the data is exchanged via the buffer 12a.

In this embodiment, since the semiconductor memory 12c that stores the block replacement information is a high-speed memory, replacement processing can be performed quickly. In addition, since alternative processing is performed for each block, information can be input and output at high speed.

Note that when controlling a magnetic disk device whose memory is arranged so that one block consists of one sector, the sector substitution information is read from the magnetic disk device and stored in the semiconductor memory, and then Sector replacement processing may be performed in the same manner.

<Effects of the Invention> As specifically explained above in conjunction with the embodiments, according to the present invention, alternative information is stored in the semiconductor memory, and the stored alternative information is referred to to perform alternative processing. There is no need to refer to the magnetic disk device, resulting in improved data input/output efficiency.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is an explanatory diagram explaining block replacement information, FIG. 3 is a block diagram showing an example of a conventional computer system, and FIG. 4 is a block diagram of a magnetic disk device. 5 is an explanatory diagram illustrating the concept of cylinders and tracks; FIG. 6 is an explanatory diagram illustrating the memory configuration of the magnetic disk device; and FIG. 7 is an explanatory diagram illustrating sector replacement information. be. In the drawings, 11 is a host computer, 12 is a disk control device, 12a is a data buffer, 12b is a processor, 12c is a semiconductor memory, and 13 is a magnetic disk device.

Claims (3)

[Claims] (1) When a data input/output request is received from a computer, data input/output is controlled for a magnetic disk device that has a defective information area in a part of the storage area where normal data input/output is not guaranteed. In a disk control device, information area replacement information for replacing a defective information area with a replacement information area is read from a magnetic disk device during operation, stored in the built-in semiconductor memory, and the stored information area replacement information is referenced. By this,
The computer determines whether or not the information area requested for input/output is a defective information area, and if the information area requested for input/output is a normal information area, data is input/output to this information area, and the input/output request is made. A disk control device characterized in that when an information area is a defective information area, data is input/output to an alternative information area indicated by information area alternative information. (2) In claim 1, the defective information area is a defective sector, the alternative information area is an alternative sector, the information area alternative information is sector alternative information, the information area is a sector, and the normal information area is a normal sector. A disk control device featuring: (3) In claim 1, the defective information area is a defective block, the alternative information area is a substitute block, the information area substitute information is block substitute information, the information area is a block, and the normal information area is a normal block. A disk control device featuring: