JPH09282603A - Disk device and monitor value preservation control method in the same device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the reduction of a performance by a wasteful monitor value preservation processing. SOLUTION: When an even becoming a monitoring object is generated, a CPU 10 increments concerned monitor values being in a monitor value storage area 121 secured in a RAM 12 and also sets a flag 122 in order to indicate that the change of monitor values is generated. Moreover, when the request of a monitor value preservation processing is occurred in this device or the outside of the device and the CPU 10 receives it, the CPU 10 checks whether a flag is set or not and when the flag is set, the CPU 10 executes the monitor value preservation processing preserving the content of the monitor value storage area 121 in the monitor value preserving area secured in the system area of a disk 1 by regarding that either of monotor values in the monitor value storage area 121 is changed. In the contrary, when the flag is not set, the CPU 10 progresses to a next processing without executing the monitor value preservation processing.

Description

Detailed Description of the Invention

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a disk device having a failure prediction function and a monitor value storage control method in the disk device.

[0002]

2. Description of the Related Art In recent years, a magnetic disk device having a function of predicting a device failure in advance (failure prediction function) has been developed. This failure prediction function monitors an event (which has a deep causal relationship with the failure) in the magnetic disk device, and predicts the device failure from the monitored value. The monitor value is data for judging whether or not the device is in a dangerous state. Typical examples are energization time, motor start / stop (Start / Stop) times, alternative processing (defective). (Process to substitute a spare sector on a spare track prepared in advance when a sector is detected)
The number of times.

The above-mentioned monitor value changes at various times such as when the apparatus is started up and when reading / writing. This type of monitor value is generally stored in a high-speed volatile memory such as a RAM at first in order to speed up the update of the monitor value. In this state, the monitor value in the volatile memory is lost when the power is turned off or the device is reset. Therefore, before that, a monitor value saving process for saving the monitor value in the volatile memory on the non-volatile memory or the disk is required.

Since the failure prediction in the magnetic disk device is performed based on the monitor value as described above, it is necessary to surely save the change in the monitor value. For that purpose, it is necessary to frequently save the monitor value.

This monitor value is generally saved by overwriting. In this monitor value saving process, even if the monitor value to be saved is the same as the value already saved in the non-volatile memory or the disk, the same value is saved (by overwriting).

[0006]

As described above, in the prior art, the monitor value to be saved is
Even if it is the same as the value already stored in the non-volatile memory or the disk, the wasteful process of storing the monitor value is performed. Moreover, during this useless processing, other processing cannot be performed, resulting in deterioration of performance. This problem will be described with reference to the flowchart of FIG.

First, in the magnetic disk device, when a command (monitor value saving command) requesting the monitor value saving process is issued from the host device, the request for the monitor value saving process is accepted (step S11). This request acceptance is not limited to when the monitor value save command is issued from the host device, but the magnetic disk device is in a power save state (standby state called standby mode), for example.
When entering, it is performed by the function of the automatic save function of the monitor value that the device has.

In the magnetic disk device, when the monitor value storage processing request is received, the monitor value stored in the volatile memory is stored in (the monitor value storage area secured in the nonvolatile storage means such as the nonvolatile memory or the disk). In step S12, the monitor value saving process of saving the value in step 1) is started.

Here, during the monitor value storing process,
It is assumed that the host device has issued a command for disk read / write, which is the original function of the disk device (step S13).

In such a case, although the command itself from the host device is accepted, the command is received only after the monitor value saving process is completed (step S14) (or after the interruptable process is performed). The process (step S15) cannot be executed. Therefore, the performance of the disk device (drive) is degraded. Moreover, the monitor value is saved even if the monitor value to be saved is the same as the already saved value, which further deteriorates the performance.

The present invention has been made in consideration of the above circumstances, and an object thereof is to perform monitor value storage processing only when there is a change in the monitor value even if a request for monitor value storage processing occurs. By doing so, it is an object of the present invention to provide a disk device and a monitor value saving control method in the same device that can prevent performance degradation due to useless monitor value saving processing.

[0012]

SUMMARY OF THE INVENTION The present invention provides a disk device having a failure prediction function of monitoring a predetermined event which may be related to a failure in the device and predicting the failure of the device from the monitored value. Volatile storage means having a monitor value storage area for temporarily storing a value, and flag information for storing flag information indicating whether or not the contents of the monitor value storage area have changed A storage unit, a non-volatile storage unit in which a monitor value storage area for storing the monitor value is secured, and the monitor value in the monitor value storage area is updated by detecting the predetermined event. Monitor means for changing the contents of the monitor value storage area to the flag information when the monitor value of the monitor value storage area is updated by the monitor means. The second state indicating that the contents of the monitor value storage area have not changed when the monitor value of the monitor value storage area is stored in the monitor value storage area. The flag information operating means to be set to and the flag information is referred to when the monitor value saving process is requested from inside or outside the device, and only when the flag information is set to the first state. And a monitor value storage control means for storing the monitor value in the monitor value storage area in the monitor value storage area.

In this disk device, when a predetermined event is detected and the monitor value in the monitor value storage area in the volatile storage means is updated, the flag information is set to the first state. Indicates that the contents of the monitor value storage area have changed.

Therefore, when the monitor value storage processing is requested inside or outside the device, it is simply referred to the flag information to determine whether the contents of the monitor value storage area are changed, that is, the monitor value storage. Whether the monitor value of the area is not the same as or the same as the value stored in the monitor value storage area (without reading the contents of the monitor value storage area and comparing with the monitor value of the monitor value storage area)
I can judge.

Then, only when it can be determined from the flag information that the contents of the monitor value storage area have changed (when the flag information is in the first state), the monitor value in the monitor value storage area is stored as the monitor value. The monitor value saving process for saving in the area is performed. On the other hand, when it is determined that the contents of the monitor value storage area have not changed (when the flag information is in the second state), the monitor value of the monitor value storage area is stored in the monitor value storage area. The monitor value saving process is suppressed,
This can prevent performance degradation.

The monitor value storage control function of the disk device described above is used for the magnetic disk device, the magneto-optical disk device, the C
Not limited to disk devices such as D-ROM devices, it can be applied to all information devices that have a failure prediction function that monitors a predetermined event that can be related to a failure in the device and predicts the device failure from the monitored value. Is.

[0017]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a magnetic disk drive according to one embodiment of the present invention.

In FIG. 1, reference numeral 1 denotes a disk (magnetic disk) as a medium on which data is recorded, and 2 denotes data writing (data recording) to the disk 1 and data reading (data reproduction) from the disk 1. This is the head (magnetic head) used. The head 2 is provided corresponding to each data surface of the disk 1.

A large number of concentric tracks are formed on both surfaces of the disk 1, and each track is used for positioning control and the like (cylinder data indicating a cylinder number and a positional error in the cylinder indicated by the cylinder data indicating a waveform error). A plurality of servo areas on which servo data (including burst data for indicating amplitude) are recorded are arranged at equal intervals.
These servo areas are radially arranged on the disk 1 over the respective tracks from the center. The area between the servo areas is a user area. One servo area is followed by one user area to form one servo sector. In the user area of each servo sector,
A plurality of data sectors are set.

A predetermined area on each surface of each disk 1 is assigned a system area. This system area is set in a different area from the data area, and is used for storing information (defect list) of defects on the corresponding disk surface, password information, and the like.

In a predetermined area (predetermined sector) of each system area, a monitor as shown in FIG. 2 for storing monitor values of various predetermined events which may be related to a failure in the magnetic disk device. The value storage area 110 is reserved. The monitor value stored in the monitor value storage area 110 includes start / stop of the motor.
There are CSS (Contact Start Stop) number indicating the number of times, number of alternative processes, energization time, and the like. Here, the monitor value storage area 110 is secured in each of the system areas allocated to each surface of each disk 1.
This is because the monitor value is backed up, but if backup is not necessary, it may be secured in only one system area.

The disk 1 is a spindle motor (SPM)
3. High speed rotation. The head 2 is attached to a head moving mechanism called a carriage 4, and moves in the radial direction of the disk 1 by the movement of the carriage 4. The carriage 4 is driven by a voice coil motor (VCM) 5.

The spindle motor 3 and the voice coil motor 5 are connected to a motor driver 6. The motor driver 6 drives the motor 3 by supplying a control current to the spindle motor 3 and drives the motor 5 by supplying a control current to the voice coil motor 5. The value of the control current (control amount) is determined by calculation processing of the CPU (microprocessor) 10, and is given, for example, as a digital value.

Each head 2 is connected to a head IC 7 mounted on a flexible printed wiring board (FPC). The head IC 7 controls switching of the head 2, input / output of a read / write signal with the head 2, and the like, and includes a head amplifier 71 that amplifies an analog output read by the head 2.

The head IC 7 is connected to a read / write IC (read / write circuit) 8. The read / write IC 8 roughly has an encode / decode function for processing user data and a signal processing function for processing servo data.

The read / write IC 8 inputs the analog output (the read signal of the head 2) read from the disk 1 by the head 2 and amplified by the head amplifier 71 in the head IC 7, and a signal necessary for the data reproducing operation. Processing, for example, signal processing for converting analog output to NRZ data and transferring it to the disk controller (HDC) 14 is performed by the decoding function. Read / write IC8
Is a signal processing required for the data recording operation, for example, HDC1
The data (eg, 2-7, 1) that modulates the NRZ data (write data) sent from
-7 modulated data) and sends it to the head IC 7 by an encode function.

The read / write IC 8 also executes, by a signal processing function, a reproducing process of servo data necessary for a servo process such as head positioning control, in addition to the above-mentioned normal recording / reproducing process of user data. That is, read / write I
C8 processes the servo data in the servo area read by the head 2 and outputs a data pulse including cylinder data to the servo processing circuit 9. Also read / write I
C8 samples and holds the peak of the burst data (in the servo data) and outputs it to the servo processing circuit 9.

The servo processing circuit 9 is a read / write IC.
In response to the data pulse and the burst data from the control unit 8, signal processing necessary for servo processing is executed. That is, the servo processing circuit 9 has a decoding function for extracting and decoding cylinder data (cylinder number) and the like from the data pulse from the read / write IC 8, and a timing generation function such as a write gate. Also, the servo processing circuit 9 has a read / write I
A / D burst data (analog signal) from C8
It also has an A / D conversion function of converting (analog / digital) and outputting it to the CPU 10. The servo processing circuit 9 is configured using, for example, a gate array (GA).

The CPU 10 is, for example, a one-chip microprocessor. The CPU 10 controls each unit in the magnetic disk device according to a control program (firmware) stored in the ROM 11.

The CPU 10 controls the head 2 in accordance with the servo data (cylinder data and burst data therein) extracted by the servo processing circuit 9 (by controlling the voice coil motor 5 via the motor driver 6). Positioning control for moving to the target position, H
There is transfer control of read / write data by controlling the DC 14. In addition to this, the CPU 10 monitors various predetermined events and manages failure prediction that predicts a failure of the device from the monitored value. Further, the CPU 10 also performs a process of updating the monitor value and a process of saving the updated monitor value in the monitor value saving area 110 in order to enable the failure prediction.

The CPU 10 includes a ROM 1 as a non-volatile memory that stores a control program (firmware) for controlling each unit in the magnetic disk device.
1 is connected to a RAM 12 as a rewritable memory that provides a work area of the CPU 10, a storage area for parameters used by the CPU 10, and the like.

In a predetermined area of the RAM 12, a monitor value storage area 121 for temporarily storing the monitor value such as the CSS number and a flag 122 setting area are secured. This flag 122 indicates that the monitor value storage area 1
21 indicates whether the content of 21 has changed (in the case of flag set) or not (in the case of flag reset). When the magnetic disk device is powered on, the contents of the monitor value storage area 110 are loaded into the monitor value storage area 121, and the flag 12
2 is reset. This flag 122 is set when the monitor value in the monitor value storage area 121 is updated.

The CPU 10 is also connected to an EEPROM 13 as a rewritable non-volatile memory in which parameters for controlling the magnetic disk device are stored and a disk controller (HDC) 14.

The disk controller (HDC) 14 is
It controls command and data communication with a host device (not shown), and also controls data communication with the read / write IC 8 (via the disk 1). This H
The DC 14 is connected to a buffer memory (buffer RAM) 15 including, for example, a RAM or the like, in which read / write data is stored in a cache system, and a host interface 16. Host interface 16
Is an interface between the HDC 14 and the host device, and the HDC 14 communicates commands and data with the host device via the host interface 16.

A timer 17 is also connected to the CPU 10. The timer 17 is activated, for example, every time the command execution is completed, and is used to measure a fixed time.
If the time measurement by the timer 17 ends (timeout)
By the time the new command is not accepted,
It is set to standby mode, which puts the device in a standby state to save power. When the standby mode is set, an auto save function for automatically saving the contents of the monitor value storage area 121 in the monitor value storage area 110 operates.

Next, the operation of the configuration of FIG. 1 will be described with reference to the flowchart of FIG. 3 regarding the monitoring process of the CSS number and the monitor value storing process for storing various monitor values including the CSS number. .

Now, it is assumed that the power of the apparatus is off. When the power of the device is turned on in this state, the CPU 10
The startup process of the device is performed by the control of. The start-up process includes a motor start-up process for starting the spindle motor 3, and the start-up current is supplied from the CPU 10 to the spindle motor 3 via the motor driver 6 to start the motor 3 ( Step S1).

When the spindle motor 3 can be started when the power is turned on, the CPU 10 carries out a process of reading information on the system area of the disk 1 into the RAM 12 or the like. By this processing, for example, the contents of the monitor value storage area 110 in the system area (various monitor values including the CSS count)
Is loaded into the monitor value storage area 121 in the RAM 12. At this time, the flag 122 in the RAM 12 is in the reset (“0” clear) state, and the monitor value storage area 11
Indicates that the content of 0 has not changed (not updated).

Since the spindle motor 3 has been activated, the CPU 10 increments the CSS count, which is a monitor value related to motor start / stop, among the monitor values stored in the monitor value storage area 121 in the RAM 12 by one. (Step S2).

At the same time, the CPU 10 sets the flag 122 in the RAM 12 to the set state (here, "1" state) to indicate that the content of the monitor value storage area 121 has changed (the monitor value has been updated). (Step S
3).

The process of starting the spindle motor 3 is as follows.
Not only when the power is turned on, but when the device returns to the normal mode to execute the command (disk access indicated by the command) as a result of the host device giving the read / write command from the standby mode Is also done. Also in this case, the above steps S2, S
3 is performed, the number of CSSs in the monitor value storage area 121 is incremented, and the flag 122 is set.

Such monitor value updating and flag setting operations are similarly performed when another event to be monitored occurs, for example, when defective sector replacement processing occurs. However, when replacement processing of a defective sector occurs, the number of replacement processing operations is updated (incremented).

After that, in the magnetic disk device of FIG. 1, normal operation (various processing such as reading / writing on the disk 1) is performed according to the request from the host device. In order to surely save the monitor value in the monitor value storage area 121, a command (monitor value save command) requesting the monitor value save process from the host device to the magnetic disk device of FIG. 1 is issued. It has been issued. The monitor value save command issued from this host device is transmitted to the HDC 14 via the host interface 16.
It is received by and is passed to the CPU 10. Then CPU10
Receives a request (request for monitor value save processing) indicated by the monitor value save command from the host device (step S4).

The request for the storage processing of the monitor value is made not only from the host device (external to the magnetic disk device),
As described below, it also occurs (equivalently) inside the magnetic disk device.

In the present embodiment, the CPU 10 starts the timer 17 when the execution of the command from the host device is finished, and whether or not a new command is issued from the host device within the fixed period measured by the timer 17. To monitor. If a new command is received within this time measurement period, the CPU 10 resets the timer 17 and executes the command.

On the other hand, if a new command is not accepted even after the time measurement of the timer 17 is completed, that is, if the time-out occurs, the CPU 10 enters the standby mode in which the device is in a standby state for power saving. Set. In this standby mode, the spindle motor 3
Is stopped.

When the standby mode is set, the contents of the monitor value storage area 121 are changed to the monitor value storage area 11
The auto-save function for automatically saving to 0 operates, and the process is performed as if a request to save the monitor value was issued inside the magnetic disk device. That is, even when the standby mode is set, the monitor value storage processing request acceptance processing of step S4 is performed as in the case where the monitor value storage command (requesting the monitor value storage processing) is issued from the host device.

When the CPU 10 receives a monitor value storage process request in step S4 as a result of the monitor value storage command being issued from the host device or as a result of setting the standby mode, the monitor value storage area 121 in the RAM 12 is received. RAM12 to check whether the contents of the contents have changed (since the last time the monitor value was saved).
By referring to the flag 122 therein, it is checked whether or not the flag 122 is set (step S5).

If the flag 122 is set, the CPU 10 determines that at least one of the various monitor values stored in the monitor value storage area 121 has changed (updated). Then, the monitor value saving process of saving the contents of the monitor value storage area 121 by overwriting in the monitor value saving area 110 secured in the system area of the disk 1 is started (step S6).

In the present embodiment, the monitor value storage area 110 is secured in each system area allocated to each surface of each disk 1, and the monitor value storage processing described above is performed in all the system areas. This is performed on the storage area 110.

It is assumed that a command (read / write command) for disk read / write is issued from the host device during the monitor value saving process (step S7). Although the read / write command itself from the host device is accepted by the interrupt process, the process of the read / write command cannot be immediately executed because the monitor value saving process is in progress.

When the monitor value saving process started in step S6 ends (step S8), the CPU 10
Resets the flag 122 in the RAM 12 (step S9), and then executes the next process (here, the process requested by the read / write command received during the monitor value saving process) (step S10).

On the other hand, when the flag 122 is not set (when the determination in step S5 is No), the CPU 10 changes all the monitor values stored in the monitor value storage area 121. Not (not updated), therefore monitor value storage area 12
It is determined that the content of 1 is the same as the content of the monitor value storage area 110. In this case, the CPU 10 ignores the request for the monitor value saving process accepted in step S4,
The process proceeds to the next process without performing the monitor value saving process.

As described above, when the flag 122 is not set, that is, when the contents of the monitor value storage area 121 are not changed, the monitor value saving process is not performed, so that the monitor value saving process is performed. Immediately after receiving the request, even if the read / write command is issued from the host device as in step S7, the read / write command can be promptly executed in step S10. Therefore, in response to the request for the monitor value saving process, the flag 122 indicates the performance deterioration of the magnetic disk device due to the monitor value saving which is always performed in the prior art.
Can be suppressed if is not set.

The effects of the present embodiment described above will be described in comparison with the prior art using the processing time table of FIG.
First, in the present embodiment, when the monitor value saving process is accepted, the state of the flag 122 is determined by referring to the flag 122, so that useless execution of the monitor value saving process is suppressed if the flag 122 is not set. To be done. Therefore, as shown in FIG. 4, assuming that the time required to determine the flag 122 is T2 and the time required to store the monitor value is T3, the time (T3-T2) is longer than that in the conventional technique.
You can move on to the next process as soon as possible.

However, in the present embodiment, the flag 122
When the flag 122 is set, assuming that the time required to set the flag 122 is T1, it takes an extra time (T1 + T2) as compared with the prior art. However, since T1 and T2 are extremely small times as compared with other processing, the influence can be ignored.

In the above embodiment, the latest monitor value is shown in the monitor value storage area 121 in the RAM 12, and in the monitor value storage processing, the monitor value in the monitor value storage area 121 is the system area of the disk 1. Although it is described that the monitor value storage area 110 secured in the above is overwritten, the present invention is not limited to this. For example, in the monitor value storage area 121, only the change amount of the monitor value is displayed, and in the monitor value storage processing, the change amount of the monitor value storage area 121 is changed to the monitor value stored in the monitor value storage area 110. You may make it add and may update the said monitor value.
In this case, the content of the monitor value storage area 121 is cleared to “0” each time the monitor value saving process is completed, and
The flag 122 needs to be reset. However, it is not necessary to load the contents of the monitor value storage area 110 into the monitor value storage area 121 when the device is started up.

In the above embodiment, the case where the disk 1 is used for storing and backing up the monitor value has been described. However, the EEPROM 13 may be used for storing and backing up the monitor value or backing up the monitor value. Absent. Further, the monitor value storage area 121 and the flag 122 may be prepared in the buffer memory 15 instead of the RAM 12.

Although the case where the present invention is applied to a magnetic disk device has been described above, the present invention relates to a magneto-optical disk device,
It can be applied to all disk devices such as CD-ROM devices. Further, the present invention is not limited to disk devices, and can be applied to all information devices having a failure prediction function of monitoring a predetermined event that may be related to a failure in a device and predicting a device failure from the monitored value.

[0060]

As described above in detail, according to the present invention, it is possible to easily determine whether or not there is a change in the monitor value when the monitor value saving process is requested, and the monitor value saving process is performed. Since it is configured to be executed only when there is a change in performance, it is possible to prevent performance degradation due to unnecessary monitor value storage processing.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a magnetic disk drive according to an embodiment of the present invention.

FIG. 2 is a diagram showing a monitor value storage area 110 secured in a system area of the disk 1.

FIG. 3 is a flowchart for explaining the operation in the embodiment, a process at the time of monitoring the CSS number, and a monitor value saving process for saving various monitor values including the CSS number.

FIG. 4 is a diagram for explaining the effect of the embodiment in comparison with a conventional technique.

FIG. 5 is a flowchart showing a conventional processing procedure when a monitor value saving process is requested.

[Explanation of symbols]

1 ... Disk (nonvolatile storage means), 2 ... Head, 3 ... Spindle motor (SPM), 10 ... CPU (monitor means, flag information operation means, monitor value storage control means), 11 ... ROM, 12 ... RAM (volatile) Memory storage means), 13 ... EEPROM (nonvolatile storage means), 15 ... buffer memory, 110 ... monitor value storage area, 121 ... monitor value storage area, 122 ... flag.

Claims (3)

[Claims] 1. A disk device having a failure prediction function of monitoring a predetermined event which may be related to a failure in a device and predicting a failure of the device from the monitored value, wherein the monitored value is temporarily stored. A monitor value storage area for securing a volatile storage means, a flag information storage means for storing flag information indicating whether or not the contents of the monitor value storage area have changed, and the monitor value A non-volatile storage means for securing a monitor value storage area for storing the monitor value, a monitor means for updating the monitor value in the monitor value storage area by detecting the predetermined event, A flag information indicating that the content of the monitor value storage area has changed when the monitor value of the monitor value storage area is updated by means; And the monitor value of the monitor value storage area is stored in the monitor value storage area, the flag information is set to a second state indicating that the content of the monitor value storage area has not changed. The flag information operating means to be set and the flag information is referred to when the monitor value storage process is requested from inside or outside the device, and when the flag information is set to the first state. And a monitor value storage control means for storing the monitor value in the monitor value storage area in the monitor value storage area. 2. A monitor value storage control method in a disk device having a failure prediction function of monitoring a predetermined event which may be related to a failure in a device and predicting a failure of the device from the monitor value, A monitor value storage area for temporarily storing the monitor value is secured in advance in a non-volatile storage means, and the monitor value in the monitor value storage area is updated by detecting the predetermined event. Along with that, flag information indicating that the contents of the monitor value storage area has changed is set, and when the storage processing of the monitor value is requested from inside or outside the device, it indicates that the contents of the monitor value storage area have changed. Only when the flag information is set, save processing is executed to save the monitor value in the monitor value storage area in the non-volatile storage means. , Monitor value saving control method in a disk device and changes the flag information to indicate that the contents of the monitoring value storage area has not changed. 3. An information device having a failure prediction function of monitoring a predetermined event which may be related to a failure in a device and predicting a failure of the device from the monitored value, wherein the monitor value is temporarily stored. A monitor value storage area for securing a volatile storage means, a flag information storage means for storing flag information indicating whether or not the contents of the monitor value storage area have changed, and the monitor value A non-volatile storage means for securing a monitor value storage area for storing the monitor value, a monitor means for updating the monitor value in the monitor value storage area by detecting the predetermined event, A flag information indicating that the content of the monitor value storage area has changed when the monitor value of the monitor value storage area is updated by means. The monitor value storage area is stored in the monitor value storage area, the flag information is set to a second state indicating that the content of the monitor value storage area has not changed. Flag information operating means for controlling the storage of the monitor value from inside or outside the device, and refers to the flag information only when the flag information is set to the first state. And a monitor value storage control means for storing the monitor value in the monitor value storage area in the monitor value storage area.