Method and system for proactive drive replacement for high availability storage systems
View Patent ↗Methods for preventing the failure of disk drives in storage systems are disclosed. A system and a computer program product for preventing the failure are also disclosed. Factors relating to the aging or early onset of errors in a disk drive are monitored. These factors are then compared to thresholds. In case the thresholds are exceeded, an indication for the replacement of the disk drive is given. Sudden rises in the factors are also used to indicate the impeding failure of disk drives.
1. A method for preventing disk drive failures, the method comprising;
monitoring factors relating to errors associated with a particular disk drive in the plurality of disk drives;
determining a first change between a second value of at least one factor and a first value of the at least one factor, the first value being a difference between a first measurement and a subsequent second measurement of the at least one factor, the second value being a difference between the second measurement and a subsequent third measurement of the at least one factor;
determining whether the first change of the at least one factor indicates a threshold increase;
determining a second change between a subsequent third value of the at least one factor and the second value of the at least one factor in response to the first change of the at least one factor indicating the threshold increase, the third value being a difference between the third measurement and a subsequent fourth measurement of the at least one factor;
determining whether the second change of the at least one factor indicates the threshold increase; and
indicating that the particular disk drive should be replaced in response to the second change of the at least one factor indicating the threshold increase.
2. The method of claim 1 , wherein the storage system includes power management.
3. The method of claim 1 , wherein the factors include power-on hours.
4. The method of claim 3 , wherein the threshold increase is derived from the mean-time-to-failure of the plurality of disk drives.
5. The method of claim 3 , wherein the mean-time-to-failure is derived from a function of temperature measurement.
6. The method of claim 3 , wherein the threshold increase includes a fraction of the mean-time-to-failure used as a threshold value to which the number of power-on hours is compared.
7. The method of claim 1 , wherein the factors include start stops.
8. The method of claim 7 , wherein the threshold increase includes a fraction of the maximum value of start stops used as a threshold value to which the start stops is compared.
9. The method of claim 1 , wherein the factors include a reallocated sector count.
10. The method of claim 9 , wherein the threshold increase includes a fraction of the maximum value of reallocated sector count used as a threshold value to which the reallocated sector count is compared.
11. The method of claim 1 , wherein the factors include a read error rate.
12. The method of claim 11 , wherein the threshold increase includes a fraction of the maximum value of read error rate used as a threshold value to which the read error rate is compared.
13. The method of claim 1 , wherein the factors include a seek, error rate.
14. The method of claim 13 , wherein the threshold increase includes a fraction of the maximum value of seek error rate used as a threshold value to which the seek error rate is compared.
15. The method of claim 1 , wherein the factors include a spin retry count.
16. The method of claim 15 , wherein the threshold increase includes a fraction of the maximum value of spin retry count used as a threshold value to which the spin retry count is compared.