Method for predicting disk drive failure by monitoring the rate of growth of defects within a disk drive

• US 5,917,724 A
• Filed: 12/20/1997
• Issued: 06/29/1999
• Est. Priority Date: 12/20/1997
• Status: Expired due to Term

• Current Assignee: Teradata US, Inc. (Teradata Corporation)
• Original Assignee: NCR Corporation
• Inventors: Denosky, James M., Bryan, William Z., Brousseau, Robert P.
• Primary Examiner(s): Barlow, John
• Assistant Examiner(s): Bui, Bryan
• Application Number:
  US08/995,432
• Time in Patent Office: 556 Days
• Field of Search: 364/186, 364/188, 364/468.15, 364/468.16, 364/468.17, 364/468.22, 364/468.23, 360/31, 360/48-49, 360/53-54, 369/32, 369/54, 369/58
• US Class Current: 700/81
• CPC Class Codes: G11B 19/04 Arrangements for preventing, inhibiting, or warning against double recording on the same blank or against other recording or reproducing malfunctions
• CPC Class Codes: G11B 20/1816 Testing
• CPC Class Codes: G11B 27/36 Monitoring, i.e. supervising the progress of recording or reproducing for digital recording G11B20/00 and s.gr.; for monitoring, testing or measuring of TV recorders of the type covered by H04N5/76 and subgroups, see H04N17/06

Abstract

A method for predicting the failure of a hard disk drive within a computer system and replacing the hard disk drive prior to an in-service failure of the disk drive. The method includes the steps of: establishing a maximum linear growth rate for the grown defect list (G-List) for the disk drive, e.g., 0.01% of the disk drive'"'"'s capacity divided by the disk drive'"'"'s expected service life of five years; periodically determining the number of entries contained in the grown defect list; and calculating the actual linear growth rate of the grown defect list from the number of entries contained in the grown defect list and the length of time the disk drive has been in service. Should the actual linear growth rate of the G-List exceed the maximum linear growth rate for two successive measurement periods, the disk drive will be replaced. The method may further include the steps of: establishing a maximum burst growth rate for the disk drive, for example four times the maximum linear growth rate; and calculating the actual burst growth rate of the grown defect list from the change in the number of entries contained in the grown defect list during a measurement period. Should the actual burst growth rate of the G-List exceed the maximum burst growth rate for two successive measurement periods, the disk drive will also be replaced.

Claims

1. A method for predicting the failure of a hard disk drive within a computer system, said hard disk drive device including a grown defect list containing a listing of storage locations within said hard disk drive device identified as damaged during use of the hard disk drive device, said method comprising the steps of:

establishing a maximum growth rate for the grown defect list for said disk drive;

periodically determining the number of entries contained in said grown defect list;

calculating the actual growth rate of said grown defect list from the number of entries contained in the grown defect list and the length of time said disk drive has been in service; and

replacing said disk drive if said actual growth rate exceeds said maximum growth rate during a specified period of time.

2. The method in accordance with claim 1, wherein:

said maximum growth rate equals 0.01% of the capacity of said disk drive divided by the service life of said disk drive.

3. The method in accordance with claim 2, wherein:

the service life of said disk drive is five years.

4. A method for predicting the failure of a hard disk drive within a computer system, said hard disk drive device including a grown defect list containing a listing of storage locations within said hard disk drive device identified as damaged during use of the hard disk drive device, said method comprising the steps of:

a) establishing a maximum growth rate for the grown defect list for said disk drive;

b) determining the number of entries contained in said grown defect list;

c) calculating an actual growth rate of said grown defect list from the number of entries contained in the grown defect list and the length of time said disk drive has been in service;

d) comparing said actual growth rate with said maximum growth rate;

e) repeating steps b) through d) at predetermined time intervals; and

f) replacing said disk drive if said actual growth rate exceeds said maximum growth rate during two consecutive time intervals.

5. The method in accordance with claim 4, wherein:

said maximum growth rate equals 0.01% of the capacity of said disk drive divided by the service life of said disk drive.

6. The method in accordance with claim 5, wherein:

the service life of said disk drive is five years.

7. A method for predicting the failure of a hard disk drive within a computer system, said hard disk drive device including a grown defect list containing a listing of storage locations within said hard disk drive device identified as damaged during use of the hard disk drive device, said method comprising the steps of:

establishing a maximum burst growth rate for the grown defect list for said disk drive;

determining the number of entries contained in said grown defect list at the beginning of a predetermined measurement period;

determining the number of entries contained in said grown defect list at the conclusion of said measurement period;

calculating the actual burst growth rate of said grown defect list from the change in the number of entries contained in the grown defect list during said measurement period and the duration of measurement period; and

replacing said disk drive if said actual burst growth rate exceeds said maximum burst growth rate during a specified period of time.

8. A method for predicting the failure of a hard disk drive within a computer system, said hard disk drive device including a grown defect list containing a listing of storage locations within said hard disk drive device identified as damaged during use of the hard disk drive device, said method comprising the steps of:

a) establishing a maximum burst growth rate for the grown defect list for said disk drive;

b) determining the number of entries contained in said grown defect list at the beginning of a predetermined measurement period;

c) determining the number of entries contained in said grown defect list at the conclusion of said measurement period;

d) calculating the actual burst growth rate of said grown defect list from the change in the number of entries contained in the grown defect list during said measurement period and the duration of measurement period;

e) comparing said actual burst growth rate with said maximum burst growth rate;

f) repeating steps b) through e) for consecutive measurement periods; and

g) replacing said disk drive if said actual burst growth rate exceeds said maximum burst growth rate during two consecutive time intervals.

9. A method for predicting the failure of a hard disk drive within a computer system, said hard disk drive device including a grown defect list containing a listing of storage locations within said hard disk drive device identified as damaged during use of the hard disk drive device, said method comprising the steps of:

establishing a maximum linear growth rate for the grown defect list for said disk drive;

establishing a maximum burst growth rate for the grown defect list for said disk drive;

establishing a repeating measurement period and determining the number of entries contained in said grown defect list at the conclusion of each measurement period;

calculating the actual linear growth rate of said grown defect list at the conclusion of each measurement period, said actual linear growth rate being calculated from the number of entries contained in the grown defect list and the length of time said disk drive has been in service;

replacing said disk drive if said actual linear growth rate exceeds said maximum growth rate during two successive measurement periods;

calculating an actual burst growth rate of said grown defect list at the conclusion of each measurement period, said actual burst growth rate being calculated from the change in the number of entries contained in the grown defect during a measurement period; and

replacing said disk drive if said actual burst growth rate exceeds said maximum growth rate during two successive measurement periods.

10. The method in accordance with claim 9, wherein:

said maximum linear growth rate equals 0.01% of the capacity of said disk drive divided by the service life of said disk drive; and

said maximum burst growth rate equals 0.08% of the capacity of said disk drive divided by the duration of said measurement period.