JPH1011701A - Magnetic disk device and fault predicting method for the same device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a magnetic disk device and the fault predicting method for the same device capable of performing a correct fault diagonosis without lowering the performance. SOLUTION: A first diagonosis by the number of collections of error collection codes(ECC) is performed in a normal time (non-alarm time) and a second diagonosis by the error rate is performed in fault alarming based on the first diagonosis result. These first and second diagonosis are realized by a firmware which is coded in a ROM 10 and controls media 1, a RAM 11 and a buffer RAM 14 or the like.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic disk drive for performing data recording (or reading) on a rotating disk-shaped magnetic disk by performing a seek operation of a magnetic head, and more particularly to a magnetic disk having a failure diagnosis function. The present invention relates to a device and a failure prediction method for the device.

[0002]

2. Description of the Related Art Conventionally, this kind of magnetic disk device is widely used as an external storage device of a computer. A magnetic disk drive that performs data recording (or reading) on a rotating disk-shaped magnetic disk (hereinafter referred to as a medium) by a seek operation of a magnetic head in a radial direction of the medium has a mechanically operating part. Therefore, the failure rate is higher than that of, for example, a semiconductor device. Therefore, there has been a need for an artificial failure countermeasure, for example, in which a user periodically backs up data.

In consideration of such a situation, magnetic disk drives having a function of diagnosing a failure by the apparatus itself have been commercialized. According to such a failure self-diagnosis function in the magnetic disk device, the RAW error rate (raw error rate, hereinafter simply referred to as “error rate”), C
Parameters such as SS (Contact Start Stop) count, start-up retry, seek performance, and other parameters that indicate equipment failures are periodically checked to determine if there is a dangerous state where failures can be predicted. Is done.
Then, when it is diagnosed that there is a dangerous state, a warning is issued to the host system in advance. It is inconvenient to issue a warning to the host system if it is too early, and it is meaningless if it is too late.

[0004]

The magnetic disk drive having the self-diagnosis function for such a failure has the following problems. In other words, if a large number of parameters are checked, a more accurate failure diagnosis can be performed.
There is a problem that the data transfer capability (performance) for reading and writing data, which is an essential function of the magnetic disk device, is reduced. SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problem, and has as its object to provide a magnetic disk drive capable of performing accurate failure diagnosis without deteriorating performance and a failure prediction method for the same.

[0005]

According to a first aspect of the present invention, there is provided a magnetic disk drive, comprising: a first diagnostic function capable of diagnosing whether there is a high possibility of a failure; and a processing time shorter than the first diagnostic function. Diagnostic means having a second diagnostic function,
Determining means for determining whether or not to perform the diagnosis by the first diagnosis function, according to a diagnosis result of the second diagnosis function of the diagnosis means.

According to a fourth aspect of the present invention, there is provided a magnetic disk drive, comprising: a plurality of diagnostic means capable of diagnosing whether there is a high possibility of failure; and a diagnostic result by some of the plurality of diagnostic means. And selecting means for selecting a diagnosis by another diagnostic means based on the above.

A magnetic disk drive according to a sixth aspect of the present invention is a magnetic disk drive having a self-diagnosis function relating to failure prediction, wherein the content of diagnosis by the self-diagnosis function is changed based on the result of diagnosis by the self-diagnosis function. It is characterized by the following.

According to a seventh aspect of the present invention, there is provided a method for predicting a failure of a magnetic disk drive, wherein the first diagnostic function is capable of diagnosing whether there is a high possibility of failure and the processing time is shorter than the first diagnostic function. Diagnosing means having a second diagnostic function, wherein it is determined whether or not to perform a diagnosis by the first diagnostic function according to a diagnosis result of the second diagnostic function of the diagnostic means. .

According to a tenth aspect of the present invention, there is provided a failure prediction method for a magnetic disk drive, wherein a plurality of diagnostic means capable of diagnosing whether or not there is a high possibility of failure, and some diagnostic means among the plurality of diagnostic means. The diagnosis by another diagnosis means is selected based on the diagnosis result by the means.

A magnetic disk drive failure prediction method according to a twelfth aspect of the present invention is a magnetic disk drive having a self-diagnosis function related to failure prediction. The content is changed.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. (First Embodiment) FIG. 1 is a block diagram showing a hardware configuration of a magnetic disk drive according to a first embodiment of the present invention. In the magnetic disk device shown in the figure, a plurality of media (here, two media) 1 are provided so as to be able to rotate at high speed by a spindle motor 17. A pair of heads 2 facing each other is provided on each data surface of the medium 1. In the case of one media 1,
Two heads 2 are provided to face both sides of the medium 1.

The head 2 is mounted on a head moving mechanism 3 called a carriage. The head 2 is movable in the radial direction of the medium 1 by the operation of the carriage 3. The positioning of the head 2 is controlled on a specified target cylinder (target track) by a servo processing system (head positioning control mechanism). Normally, data is read / written in sector units.

The carriage 3 is provided with a voice coil motor (V
CM) 16 and the VCM 16 is a driver 15
Driven by The spindle motor 17 is also driven by the driver 15. The driver 15 is a driver IC (integrated circuit) constituting a motor driver and a VCM driver (referred to as a double driver).

The read / write circuit 8 includes a head amplifier 4
Input the read signal from the head 2 amplified by
Performs signal processing necessary for data reproduction operation. The read / write circuit 8 performs signal processing necessary for the data recording operation, and supplies a write current corresponding to the write data to the head 2 via the head amplifier 4.

The read / write circuit 8 executes a recording process of the user data and a reproducing process of the servo data necessary for the servo process of the head positioning control, in addition to a normal recording process of the user data. The servo data includes a cylinder address indicating the current position of the head 2 and burst data (burst signal) for indicating a position error in the cylinder. Servo circuit 5
Is a circuit for executing signal processing necessary for servo processing, and extracts and holds a cylinder address from a data pulse output from the read / write circuit 8. Further, the servo circuit 5 outputs a sample timing signal for extracting a burst signal to the pulse generation circuit 6 or performs a process of extracting a sector pulse.

The CPU 9 constitutes a servo processing system for executing head positioning control together with the servo circuit 5.
Further, the CPU 9 executes various drive controls of the magnetic disk drive, including read / write data transfer control, in addition to head positioning control. The CPU 9 executes various drive controls based on a control program stored in the ROM 10. The ROM 10 is connected to the CPU 9 via an external bus. The ROM 10 stores firmware for failure diagnosis of the magnetic disk device. The ROM 10 also stores conditions such as a change in the output level of data, the number of repeated read processes (the number of retries), and the fact that data was correctly read by an error correction function such as ECC.

The CPU 9 is a one-chip microprocessor, and has an A / D converter and a D /
A built-in A converter and an internal read / write memory called an instruction RAM. In the embodiment,
The CPU 9 converts the supplied burst signal BS into digital data using an A / D converter. Also, CP
U9 uses a D / A converter to convert a control amount required for servo processing (a control amount required for head positioning control and the like) into an analog signal and output the analog signal to the driver 15.

The HDC 12 constitutes an interface between the magnetic disk drive and the host system, and is a controller for mainly transferring read / write data.
The HDC 12 temporarily stores in the buffer RAM 14 read data read from the medium 1 in sector units and write data for writing to the medium 1. HD
The C12 and the CPU 9 constitute a system for performing data transfer control. Further, the HDC 12 is provided with an error detection circuit. This error detection circuit is used for media 1
In addition to detecting a data error based on the data read out from the memory, it has a function of recovering the error in a predetermined narrow area, that is, an ECC (Error Correction Code) function. The EEPROM 7 is connected to the CPU 9,
Various types of data are stored.

The operation of the magnetic disk drive according to the embodiment configured as described above will be described. FIG. 2 is a flowchart showing the operation of the magnetic disk drive according to the embodiment. The operation described here is realized by firmware stored in the ROM 10.

First, in step S0, an initialization process is performed. This step S0 is executed immediately after the power of the magnetic disk drive is turned on. In the initialization process, HDC12,
The read / write circuit 8, the alert mode flag, and the like are initialized (reset). Then, control goes to a step S2.

Step S2 is an idle routine for waiting for a command from the host system. When a command is received from the host system, a command interrupt is issued, and the flow shifts to step S10 (command processing).

In the idle state in step S2, if no command arrives from the host system for a certain period of time (for example, 5 seconds), the process proceeds to step S4. In step S4, it is determined based on the alert mode flag whether or not a failure alert is on. The alert mode flag is RA
This is a flag stored and held in M11. If a failure alert is being made, the process proceeds to step S6. In step S6, a high-precision failure diagnosis that can accurately predict the occurrence of a failure, here, measurement of an error rate, is performed. Although the error rate is an effective parameter indicating a factor leading to a failure of the device, a relatively long measurement time (several tens of seconds or more) is required because it is a data measurement including a statistical method. After the measurement of the error rate, the process shifts to a power down process in the power save mode.

If no alert is issued, the process simply shifts to the power down process without executing step S6. As a result, it is possible to suppress a decrease in performance due to the measurement of the error rate in a normal state in which no alert is issued. The description of the power down processing is omitted here.

When a command is received from the host system during the measurement of the error rate, the measurement may be interrupted to execute the command, or the command execution may be delayed until the measurement is completed. Delaying the execution of a command causes performance degradation. Even when the measurement is interrupted, performance is reduced due to overhead. Therefore, the error rate measurement should not be performed frequently.

However, the present embodiment is preferable because the error rate measurement is performed only under specific conditions (during alert). FIG. 3 is a flowchart showing the command processing (S10).

First, in step S11, the type of command is determined based on the command code. Next, when it is determined in step S12 that the command is a “read command”, the process proceeds to a read command process in step S20.

If it is determined in step S13 that the command is "drive status read command", the process proceeds to the drive status read process in step S30.

The command processing S10 is for processing "other commands" other than the above, but the description is omitted here. FIG. 4 is a flowchart showing the read command processing (S20).

First, in step S21, a read process is performed by operations of the read / write circuit 8 and the HDC 12 according to the command sent from the host system.

Next, in step S22, a diagnosis is performed so as not to significantly affect the performance of the magnetic disk device and to approximate the error rate fluctuation characteristics, that is, to obtain the number of ECC collections. The number of ECC collections is E
This is the number of data (collected data) collected in the read processing when the CC correction is enabled. Then, control goes to a step S221. In step S221, it is determined whether or not a predetermined time has elapsed after entering the alert mode. Here, if the determination is YES, the process moves to step S23, and the alert mode flag is reset. The alert mode flag is stored and held in the RAM 11.

Next, in step S24, it is determined whether or not the number of collected data (the number of ECC collections) is larger than a predetermined threshold. Here, if the threshold value is larger than the predetermined threshold (specifically, the on-the-fly collection is performed once every 10 ³ ), the alert mode flag is set in step S25. As a result, the magnetic disk device enters a failure alert state.

FIG. 5 is a flowchart showing the drive status read command processing (S30). First, in step S31, it is determined whether or not the apparatus is in the alert mode (fault alert) by checking the alert mode flag. The alert mode flag is stored and held in the RAM 11. If the mode is the alert mode, the process proceeds to step S32. If the mode is not the alert mode, the process proceeds to step S34.

In step S32, a failure diagnosis is performed based on the measurement result of the error rate in step S6. In the following step S33, when the error rate is larger than a predetermined threshold (for example, when the error rate is 10 ^-3 or more), it is determined to be "abnormal",
Move to step S36. If the error rate is smaller than the threshold, the process moves to step S34.

In step S34, a failure diagnosis based on the number of ECC collections is performed. Subsequent step S
At 35, if the number of ECC collections is larger than a predetermined threshold (for example, on-the-fly collection is performed once every 10 ³ ), “abnormal”
And the process moves to step S36. Also the above EC
If the number of C collections is smaller than the threshold, the process moves to step S37.

In step S36, the host system 13 is notified via the host interface 13 that the drive status is abnormal. In step S37, the drive status is reported to the host system via the interface 13 that the drive status is normal. Step S34 and step S34
The diagnosis based on the number of ECC collections at 35 may be executed in the non-failure alert mode or when the diagnosis result using the error rate is “normal”, but may be omitted in some cases.

According to the first embodiment, the first diagnosis based on the number of ECC collections is performed in the normal state (in the non-warning mode), and during the failure warning based on the first diagnosis result, that is, in step S25. While the warning flag is being set, the second diagnosis based on the error rate is performed, so that an accurate failure prediction can be performed by the minimum necessary diagnostic processing.

Therefore, the host system to which the magnetic disk device is connected can notify the user of the failure prediction of the magnetic disk device at an appropriate timing.
This allows the user to take appropriate countermeasures.

The first diagnosis is not limited to the diagnosis based on the number of ECC collections, but uses other parameters that do not significantly affect the performance of the magnetic disk drive and approximate the error rate fluctuation characteristics. May be.

The diagnostic parameter for diagnosing a failure with high accuracy during failure alerting is not limited to only one type of error rate. In the present embodiment, the error rate is combined with the number of ECC collections giving its approximate characteristics. For example, the first diagnosis is a measurement of the CSS activation time, and the second diagnosis is the CSS activation current value and the number of retries. Such a combination of other diagnostic parameters may be used. That is, in the case of the first embodiment,
The first diagnosis may be a diagnosis that does not significantly affect the performance of the magnetic disk device, and the second diagnosis may be a high-precision diagnosis that can accurately predict the occurrence of a failure.

(Second Embodiment) The second embodiment relates to a modification of the first embodiment, and a description of the same parts as in the first embodiment will be omitted.

In the first embodiment, a first diagnosis is performed based on the number of ECC collections in a normal state (in a non-warning mode), and a second diagnosis is performed based on an error rate during a warning based on a result of the first diagnosis. However, in the second embodiment, a failure alert level (failure alert level) is determined according to the first diagnosis result, and various types of diagnosis (second alert) corresponding to the level are determined during the alert alert. Diagnosis). That is, in the first embodiment, the failure alert level is 1, but in the second embodiment, a plurality of failure alert levels are provided. In other words, the second diagnosis can be selected based on the diagnosis result of the first diagnosis.

According to the specific example of the second embodiment, the number of ECC collections, that is, the number of data (collected data) collected in the read process when the HDC 12 is enabled for ECC collection is used. Based on one diagnosis result, for example, three levels of failure alert levels are determined. Steps related to the failure alert level determination are replaced with steps S24 and S25 in FIG. Then, at the failure alert level 1, diagnosis based on the error rate is performed, and at the failure alert level 2, the error rate and the CSS are determined.
Diagnosis using (starting current, number of retries), and at failure alert level 3, diagnosis using error rate and CSS and seek performance.
Steps related to these diagnostic processes are performed in step S in FIG.
4 and step S6. The diagnosis of “abnormal” at the failure alert level 2 and the failure alert level 3 means a case where any one of the parameters indicates an abnormality at that level or a case where all the parameters indicate an abnormality.

Note that the first diagnosis is not limited to the diagnosis based on the number of ECC collections, but is a diagnosis that does not significantly affect the performance of the magnetic disk drive, such as the measurement of the CSS startup time, as in the first embodiment. It may be. Further, the failure alert level of the second diagnosis is not limited to three levels, and may be any level. Further, the type of the second diagnosis is not limited to the above, and any diagnosis may be performed so that the required diagnosis accuracy is obtained at a minimum.

According to the second embodiment, there is an advantage that a more flexible and diverse failure diagnosis can be performed with higher accuracy than the first embodiment. The present invention is not limited to the above-described embodiment, and can be implemented in various modifications, for example, by combining the first embodiment and the second embodiment.

[0045]

According to the present invention, only the necessary minimum self-diagnosis is performed in the initial stage where no sign of failure is observed, thereby suppressing a decrease in performance, and in the warning stage where a sign of failure is observed. Can extend the self-diagnosis according to the degree of danger, and thereby provide a magnetic disk device and a failure prediction method for performing the diagnosis with high accuracy.

[Brief description of the drawings]

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

FIG. 2 is a flowchart showing the operation of the magnetic disk drive according to the first embodiment.

FIG. 3 is a flowchart illustrating command processing of the magnetic disk device according to the first embodiment.

FIG. 4 is a flowchart showing a read command process of the magnetic disk device according to the first embodiment.

FIG. 5 is a flowchart showing a drive status read command process of the magnetic disk device according to the first embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Media (magnetic disk) 2 ... Magnetic head 3 ... Carriage 4 ... Head amplifier 5 ... Servo circuit 6 ... Pulse generation circuit 7 ... EEPROM 8 ... Read / write circuit 9 ... CPU 10 ... ROM 11 ... RAM 12 ... HDC 13 ... Host interface 14 Buffer RAM 15 Driver 16 VCM 17 Spindle motor

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification number Agency reference number FI Technical display location G11B 20/18 572 G11B 20/18 572F

Claims (12)

[Claims] A diagnostic means having a first diagnostic function capable of diagnosing whether there is a high possibility of a failure and a second diagnostic function having a shorter processing time than the first diagnostic function; Determining means for determining whether or not to perform a diagnosis by the first diagnosis function according to a diagnosis result of the second diagnosis function. 2. The magnetic disk drive according to claim 1, wherein the first diagnosis function diagnoses a possibility of a failure by measuring an error rate. 3. The magnetic field according to claim 1, wherein the diagnosis of the possibility of failure by the first diagnosis function has a higher diagnosis accuracy than the diagnosis of the possibility of failure by the second diagnosis function. Disk device. 4. A plurality of diagnosing means capable of diagnosing whether there is a high possibility of a failure, and diagnosing by other diagnosing means based on a diagnosis result by some of the plurality of diagnosing means. A magnetic disk drive, comprising: selecting means for selecting. 5. The magnetic disk drive according to claim 4, wherein the selection unit changes the number of diagnosis units to increase diagnosis accuracy. 6. A magnetic disk drive having a self-diagnosis function related to failure prediction, wherein the content of diagnosis by the self-diagnosis function is changed based on a diagnosis result by the self-diagnosis function. 7. A diagnostic device comprising: a first diagnostic function capable of diagnosing whether there is a high possibility of a failure; and a diagnostic means having a second diagnostic function having a shorter processing time than the first diagnostic function. A failure predicting method for a magnetic disk drive, comprising: determining whether to perform a diagnosis by the first diagnosis function according to a diagnosis result of a second diagnosis function of the means. 8. The method according to claim 7, wherein the first diagnosis function diagnoses a possibility of a failure by measuring an error rate. 9. The magnetic apparatus according to claim 7, wherein the diagnosis of the possibility of failure by the first diagnosis function has a higher diagnosis accuracy than the diagnosis of the possibility of failure by the second diagnosis function. Disk device failure prediction method. 10. A plurality of diagnosing means capable of diagnosing whether there is a high possibility of failure, and diagnosing by other diagnosing means based on a diagnosis result by some of the plurality of diagnosing means. A method for predicting a failure of a magnetic disk drive, comprising: 11. The failure prediction method according to claim 10, wherein the selection unit changes the number of diagnosis units to increase diagnosis accuracy. 12. A magnetic disk drive having a self-diagnosis function related to failure prediction, wherein the content of diagnosis by the self-diagnosis function is changed based on a diagnosis result by the self-diagnosis function. Method.