JPH05181612A - Disk device - Google Patents

Abstract

PURPOSE:To attain the duplex recording of information with a single device by writing simultaneously the same data into plural areas with plural heads and then reading out the data with a single corresponding head in a reproduction mode. CONSTITUTION:Plural magnetic disks 2 are fixed to the rotary shaft of a spindle motor 1. The lowest side of each magnetic disk 2 serves as a servo surface where the stored data are recorded. Then two magnetic heads 3 are provided on each side of the disk 2 so that the data are simultaneously written into the areas A and B via both heads 3. Meanwhile one of both heads 3 usually reads out the data. When one of both heads 3 has an error and is unable to read out the data, the other head 3 reads out the data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk device for storing information such as a magnetic disk device and an optical disk device.

[0002]

2. Description of the Related Art A system that can be operated without stopping operation even if some of the devices have failed is called a fault tolerant system.

In computer systems, various disk devices such as magnetic disk devices and optical disk devices are often used. When configuring a fault tolerant system with such a computer system, in general,
Multiplexing of devices is performed such that two disk devices are provided.

[0004]

However, in such multiplexing, the number of devices increases, the system becomes large, and the hardware cost rises. In addition, since it is necessary to switch the driving state such that one of the two devices is driven during normal times and the other is driven during an abnormal condition, the software cost of the operating system and the like also rises.

As described above, conventionally, since the fault tolerant system is constructed by multiplexing the devices, there is a problem that the system becomes large and the system cost rises.

It is an object of the present invention to solve the above problems and to provide a disk device which can downsize the system and reduce the system cost.

[0007]

Therefore, according to the present invention, at the time of information recording, the same data is simultaneously written in a plurality of areas on the information recording surface of the disk by using a plurality of heads, while the information is reproduced. In this case, the data is read from one of the areas in which the same data is written by one corresponding head.

[0008]

Since the information recording is duplicated in one disk device, the number of devices does not increase when the system is constructed, the system is downsized, and the hardware cost is reduced. Further, since only one device has to be driven at any time, the software cost is also reduced.

[0009]

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a schematic side view of the inside of a magnetic disk device according to an embodiment of the present invention. In the figure, a plurality of magnetic disks 2 are fixed to a rotary shaft of a spindle motor 1. Among the respective surfaces of the magnetic disk 2, the lowermost one surface is a servo surface on which servo information is recorded, and the other surfaces are data surfaces on which stored data is recorded. Two magnetic heads 3 are arranged on each data surface. The two magnetic heads 3 are connected by an arm 4, and the arm 4 is fixed to a carriage member 5. The carriage member 5 is driven by the seek motor 6.

The magnetic heads 3 are arranged at regular intervals on the information recording surface of the magnetic disk 2, as shown in FIG.
One of the magnetic heads 3 accesses a half area A of the information recording surface by the predetermined seek operation, and the other head B of the magnetic head 3
To access.

FIG. 3 is a block diagram of the control circuit of the magnetic disk device of this embodiment. In the figure, a seek control unit 101 controls the seek motor 6 to control the magnetic head 3
Is moved to a predetermined track position. The data recording / reproducing unit 102 inputs / outputs data to / from the host computer and writes / reads data to / from the magnetic disk 2. The magnetic head 3 is connected to the data recording / reproducing section 102 via an arm selecting section 103 and a head selecting section 104.
The arm selection unit 103 selects an arm used for writing and reading data. Head selection unit 1
04 selects one or both of the two magnetic heads 3 arranged on one selected arm.

The control unit 105 controls each of the above units according to a control command on the host computer side.

Next, the recording operation of the magnetic disk device of the present embodiment having the above configuration will be described.

When recording information, a write command and write data are input from the host computer.
In this embodiment, writing and reading of data are performed in track units, and the write command is used to specify a logical track number to write data.

The magnetic disk drive, as shown in FIG.
When the write command is input (process 201), first,
From the logical track number specified by the write command,
The physical address for accessing the magnetic disk 2 is determined (process 202).

By the way, in this embodiment, the magnetic disk 2 is used.
One of the two areas A and B is for backup at the time of failure, and the same physical track number is given to the areas A and B. Now, for example, if there are 200 tracks in each of the areas A and B and the data surface of the magnetic disk 2 is 5, this device is controlled with 200 cylinders and 1000 logical tracks. .. In this case, the alternative track for the defective track is formed separately from the above. Figure 5
The operation of determining the physical address is shown. That is, first, an integer operation is performed to divide the designated logical track number by the number of arms 4 to calculate the quotient (process 30).
1). Then, the integer part of the quotient is taken out to be the cylinder number, and the remainder is the arm number (process 30).
2).

Then, the cylinder number and arm number are collated with the registration information of the defective track (process 30).
3). This registration information is created by a known technique and is shown in FIG.
It is stored in a list form as shown in (a). The "bad track" in the figure is a physical address of a track in which data cannot be written or read. The "alternative track" is a physical address of a track used instead of the defective track. In this example, the physical address is
As shown in (b) of the figure, "0X" indicates hexadecimal notation, and the following two characters are the cylinder number and the subsequent two characters are the arm number. Further, in this example, “0XFFFF” of a good track is a code indicating the end of registration information.

If the obtained cylinder number and arm number are registered as defective tracks by the above collation, the cylinder number and arm number of the corresponding alternative track are adopted (process 305). If it is not registered in the defective track, the obtained cylinder number and arm number are directly used as the determination result. And
This operation ends.

When the physical address is determined in this way, the magnetic head 3 is next seek-controlled to a predetermined cylinder position (step 203 in FIG. 4), and the arm 4 to be used is selected (step 204). In this case, the write data input from the host computer is recorded on one data surface of the corresponding magnetic disk 2 by using both the two magnetic heads 3 of the selected arm 4 (process 205).
As a result, the same data is recorded on one track of the area A and one track of the area B on the data surface.

At the time of this recording operation, a servo control unit (not shown) reads control information from one servo surface provided and executes seek control. At this time, a control error may occur due to a control information read error or the like.

At the time of the recording operation, the state of seek control is checked (process 206). Then, if there is no control error and data can be recorded correctly (N in step 206), then the print data is read out by the selected two heads (step 207). Then, the read data is checked (process 208).

Here, CRC (Cyclic Redu)
If the read data is correct (N in process 208) and there is no error due to the nancy check) check, the read data is compared with the written data (process 210). Then, if the two match (Y in process 210), the write operation is terminated.

On the other hand, in the above, if a control error occurs (Y in process 206), a data error occurs (Y in process 208), or read data and print data do not match (process). 210 N), the physical address that has just been accessed is registered in the defective track table described in FIG. 6 (process 211). afterwards,
Data is similarly recorded on the corresponding alternative track (to step 202).

Next, the read operation of the recording data will be described.

When reading data, a read command is input from the host computer. As shown in FIG. 7, when the read command is input (process 401), the magnetic disk device determines the physical address from the designated logical track number as described above (process 402). Then, a predetermined seek control is executed and (step 40
3) The corresponding arm 4 is selected (process 404).

Next, one of the two magnetic heads 3 of the selected arm 4 which is set in advance is used to read the recorded data from the magnetic disk 2 (process 405). Next, the control operation and the read data are checked (process 406). Here, the seek control error and CR
If there is no data error due to C check (process 406)
N), the read data is sent to the host computer (process 407), and the read operation ends.

On the other hand, if a control error or a data error occurs (Y in step 406), the other of the two magnetic heads 3 is switched to read the recording data (step 408). Then, the control operation and the read data are checked (process 409). If there is no error (N in process 409), the read data is sent to the host computer (process 407), and the read operation ends.

If an error occurs again (Y in step 409), the host computer is notified of the error (step 410) and the read operation is terminated.

As described above, in this embodiment, two magnetic heads 3 are arranged on each surface of the magnetic disk 2, and the two magnetic heads 3 simultaneously write data to the areas A and B of the magnetic disk 2. I try to write it. In general, one of the two magnetic heads 3 reads data, and when an error occurs in one of the magnetic heads 3 and data cannot be read, the other magnetic head 3 reads data. I am trying.

As described above, in this embodiment, since one magnetic disk device duplicates data recording, it is not necessary to provide two magnetic disk devices when constructing a computer system. Also, the fault tolerant system can be configured with only one unit. As a result, the hardware cost is reduced, and at the same time, only one device has to be driven, so that the software cost is also reduced.

Further, when data is read out from one of the two magnetic heads 3 and an error occurs, the other one is automatically read out, so that the device can be monitored for errors by software or the magnetic head 3 can be used. The software is simple because there is no need to switch.

Further, since the same data is recorded in two areas on one surface of the magnetic disk 2, even if one area becomes unusable due to various errors, the storage capacity of the disk device does not decrease much. Complete.

Two magnetic heads 3 are arranged on one data surface of the magnetic disk 2 so as to be duplicated. However, if three or more magnetic heads are arranged, it is possible to further multiplex.

As shown in FIG. 7, the magnetic head 3
One may be provided for each data surface, and the same data may be recorded on one surface and the other surface of one magnetic disk 2, for example, so as to be duplicated.
This allows conventional hardware to be used as the device of the present invention.

Further, in the above-described embodiment, the number of the magnetic heads 3 on the servo surface of the magnetic disk 2 is one, but two magnetic heads 3 are provided. On the other hand, if a servo information reading error occurs, the other one is used. You may read in.

Further, although the magnetic disk device has been described as an example, the present invention can be similarly applied to various optical disk devices.

[0038]

As described above, according to the present invention, at the time of information recording, the same data is simultaneously written in a plurality of areas of the information recording surface by a plurality of heads, while at the time of information reproduction, the same data is written. Since data is read from one of the areas in which each is written by one corresponding head, a fault-tolerant system can be constructed without increasing the number of devices. The cost will be reduced.

[Brief description of drawings]

FIG. 1 is a schematic side view of the inside of a magnetic disk device according to an embodiment of the present invention.

FIG. 2 is a plan view of a magnetic disk.

FIG. 3 is a block configuration diagram of a control unit.

FIG. 4 is a flowchart of a data write operation.

FIG. 5 is a flowchart of a physical address determination operation.

FIG. 6 is an explanatory diagram of registration information of defective tracks.

FIG. 7 is a flowchart of a data read operation.

FIG. 8 is a schematic side view of the inside of a magnetic disk device according to another embodiment of the present invention.

[Explanation of symbols]

 1 spindle motor 2 magnetic disk 3 magnetic head 4 arm 5 carriage member 6 seek motor 101 seek control unit 102 data recording / reproducing unit 103, 103 arm selecting unit 104, 104 head selecting unit 105 control unit

Claims (4)

[Claims] 1. A disk device in which a head accesses an information recording surface of a disk as a storage medium to write and read data, and the same data is written in a plurality of areas of the information recording surface by using a plurality of the heads. A disk device comprising: a data writing unit that writes data at the same time; and a data reading unit that reads data from one of the plurality of areas in which the same data is written by a corresponding head. 2. The disk device according to claim 1, wherein the plurality of areas for writing the same data are formed in one information recording surface. 3. The disk device according to claim 1, wherein the areas for writing the same data are formed one by one on the information recording surface. 4. The data reading means comprises head switching means for reading the same data from another area by another head when data is read by one head and an error occurs. The disk device according to claim 1.