Information memory device, failure predicting device and failure predicting method
According to one embodiment, physical position information on errors on a recording medium is acquired, physical position relationship between the errors on the recording medium is calculated based on the position information, and a failure mode related to the errors is determined based on the position relationship.
1. An information memory device comprising:
a magnetic disc;
a magnetic head provided for the magnetic disc; and
a prediction unit that predicts a failure mode related to errors on the magnetic disc based on physical position relationship between the errors,
wherein the prediction unit predicts the failure mode based on a multivariate logistic regression model with the number of the errors and a parameter related to position information on the errors as explanatory variables.
2. The information memory device of claim 1 ,
a magnetic disc;
a magnetic head provided for the magnetic disc; and
a prediction unit that predicts a failure mode related to errors on the magnetic disc based on physical position relationship between the errors, wherein
the physical position relationship between the errors is a physical distance between the errors, and
the prediction unit predicts whether the failure mode is a disc failure or a head failure based on a parameter related to the physical distance between the errors.
3. The information memory device of claim 2 , wherein the parameter related to the physical distance between the errors indicates locality or dispersibility of error distribution on the magnetic disc.
4. The information memory device of claim 2 , wherein the physical distance is a two-dimensional spatial distance or a cylinder distance.
5. The information memory device of claim 2 , wherein the prediction unit includes a maintenance unit that, in the case of the disc failure, disables writing into a predetermined range including sites where the errors have occurred, and in the case of the head failure, migrates data to a spare region.
6. The information memory device of claim 2 , comprising a display unit that displays information including at least one of the physical position information on the errors and the parameter related to the physical distance between the errors.
7. The information memory device of claim 1 , wherein
the parameter related to the physical distance is given by a dispersive power D, and
the dispersive power D is given by the following equation:
D=W/W e
where
W
=
1
n
∑
i
=
1
n
d
i
W
e
=
1
2
n
/
S
S: Area of error distribution
di: Distance from error i to closest adjacent error
n: Number of errors.
8. The information memory device of claim 1 , wherein
the physical position relationship between the errors is radial magnitudes of the errors, and
the prediction unit predicts whether the failure mode is a write failure or a read failure based on a parameter related to the magnitudes in the radial direction in which the errors exist.
9. The information memory device of claim 8 , wherein the parameter related to the radial magnitudes of the errors is the total number of tracks with the errors, the maximum value of number of tracks with the continuous errors, or a standard deviation in the radial direction of the errors.
10. The information memory device of claim 8 , wherein the prediction unit includes a maintenance unit that disables writing in the case of the write failure and migrates data to a spare region in the case of the read failure.
11. The information memory device of claim 8 , comprising a display unit that displays information including at least one of the physical position information on the tracks with the errors and the total number of the tracks with the errors.
12. A failure predicting device comprising:
an acquisition unit that acquires physical position information on errors on a recording medium; and
a determination unit that calculates physical position relationship between the errors on the recording medium based on the position information and determines a failure mode related to the errors based on the position information,
wherein
the physical position relationship between the errors is a physical distance between the errors, and
the determination unit determines whether the failure mode is a medium failure or a head failure based on a parameter related to the physical distance between the errors.
13. The failure predicting device of claim 12 , wherein the parameter related to the physical distance between the errors indicates locality or dispersibility of error distribution on the recording medium.
14. The failure predicting device of claim 12 , wherein
the physical position relationship between the errors is radial magnitudes of the errors, and
the determination unit determines whether the failure mode is a write failure or a read failure based on a parameter related to the magnitudes in the radial direction in which the errors exist.
15. A failure predicting method comprising:
acquiring physical position information on errors on a recording medium; and
calculating physical position relationship between the errors on the recording medium based on the position information and determining a failure mode related to the errors based on the position relationship,
wherein
the physical position relationship between the errors is a physical distance between the errors, and
it is determined whether the failure mode is a medium failure or a head failure based on a parameter related to the physical distance between the errors.
16. The failure predicting method of claim 15 , wherein the parameter related to the physical distance between the errors indicates locality or dispersibility of error distribution on the recording medium.
17. The failure predicting method of claim 15 , wherein
the physical position relationship between the errors is radial magnitudes of the errors, and
it is determined whether the failure mode is a write failure or a read failure based on a parameter related to the magnitudes in the radial direction in which the errors exist.