Method, device and computer product for predicting disk failure
Embodiments of the present disclosure provide method, device and computer product for predicting disk failure. The method disclosed herein comprising: obtaining operation data of a disk, each data item of the operation data indicating values for one or more attributes of the disk at an associated time point; identifying null values for the one or more attributes from the data items of the operation data; adjusting the operation data based at least on the identifying of the null values; and processing the adjusted operation data with a machine learning model, to obtain a failure prediction on whether the disk will fail within a predetermined time period after the associated time point.
1. A method for predicting disk failure, comprising:
obtaining operation data of a disk, each data item of the operation data indicating values for one or more attributes of the disk at an associated time point;
identifying null values for the one or more attributes from the data items of the operation data;
adjusting the operation data based at least on the identifying of the null values, the adjusted operation data including a difference between a value for each attribute of the data item at a first time and a value for the attribute of the data item at a second time; and
processing the adjusted operation data with a machine learning model, to obtain a failure prediction on whether the disk will fail within a predetermined time period after the associated time point.
2. The method of claim 1 , wherein adjusting the operation data comprises:
determining a proportion of null values identified in a first data item of the operation data; and
in response to the proportion of the null values exceeding a proportion threshold, removing the first data item from the operation data.
3. The method of claim 1 , wherein adjusting the operation data comprises:
setting the null values identified in a data item of the operation data to a predetermined value.
4. The method of claim 1 , wherein adjusting the operation data further comprises:
determining, based on the operation data, an attribute variation value indicating a particular attribute of the disk, the attribute variation value indicating a degree of variation at a first time point in the value for the particular attribute of the disk; and
adding the attribute variation value to a second data item of the operation data corresponding to the first time point.
5. The method of claim 1 , wherein the disk is a serial attached small computer system interface disk, the one or more attributes comprising a background media scan attribute of the disk.
6. The method of claim 1 , wherein the machine learning model is a random forest model.
7. The method of claim 1 , further comprising:
processing test operation data with the machine learning model to obtain a test failure prediction on whether the disk will fail within a predetermined time period after a respective time point corresponding to the test operation data; and
adjusting one or more hyper parameters of the machine learning model based on the test failure prediction.
8. The method of claim 7 , wherein the one or more hyper parameters comprise a length of the predetermined time period.
9. A device for predicting disk failure, comprising:
at least one processing unit;
at least one memory coupled to the at least one processing unit and having instructions stored thereon for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, cause the at least one processing unit to perform operations, the operations comprising:
obtaining operation data of a disk, each data item of the operation data indicating values for one or more attributes of the disk at an associated time point;
identifying null values for the one or more attributes from the data items of the operation data;
adjusting the operation data based at least on the identifying of the null values, including calculating a difference between a value for each attribute of the data item at a first time and a value for the attribute of the data item at a second time; and
processing the adjusted operation data with a machine learning model, to obtain a failure prediction on whether the disk will fail within a predetermined time period after the associated time point.
10. The device of claim 9 , wherein adjusting the operation data comprises:
determining a proportion of null values identified in a first data item of the operation data; and
in response to the proportion of the null values exceeding a proportion threshold, removing the first data item from the operation data.
11. The device of claim 9 , wherein adjusting the operation data comprises:
setting the null values identified in a data item of the operation data to a predetermined value.
12. The device of claim 9 , wherein adjusting the operation data further comprises:
determining, based on the operation data, an attribute variation value indicating a particular attribute of the disk, the attribute variation value indicating a degree of variation at a first time point in the value for the particular attribute of the disk; and
adding, the attribute variation value to a second data item of the operation data corresponding to the first time point.
13. The device of claim 9 , wherein the disk is a serial attached small computer system interface disk, the one or more attributes comprising a background media scan attribute of the disk.
14. The device of claim 9 , wherein the machine learning model is a random forest model.
15. The device of claim 9 , wherein the operations further comprise:
processing test operation data with the machine learning model to obtain a test failure prediction on whether the disk will fail within a predetermined time period after a respective time point corresponding to the test operation data; and
adjusting one or more hyper parameters of the machine learning model based on the test failure prediction.
16. The device of claim 15 , wherein the one or more hyper parameters comprise a length of the predetermined time period.
17. A computer product being stored on a non-transitory computer storage medium and comprising machine-executable instructions which, when executed by a processor, cause the processor to perform operations, the operations comprising:
obtaining operation data of a disk, each data item of the operation data indicating values for one or more attributes of the disk at an associated time point;
identifying null values for the one or more attributes from the data items of the operation data;
adjusting the operation data based at least on the identifying of the null values, including calculating a difference between a value for each attribute of the data item at a first time and a value for the attribute of the data item at a second time; and
processing the adjusted operation data with a machine learning model, to obtain a failure prediction on whether the disk will fail within a predetermined time period after the associated time point.
18. The computer product of claim 17 , wherein adjusting the operation data comprises:
determining a proportion of null values identified in a first data item of the operation data; and
in response to the proportion of the null values exceeding a proportion threshold, removing the first data item from the operation data.
19. The method of claim 17 , wherein adjusting the operation data comprises:
setting the null values identified in a data item of the operation data to a predetermined value.
20. The method of claim 17 , wherein adjusting the operation data further comprises:
determining, based on the operation data, an attribute variation value indicating a particular attribute of the disk, the attribute variation value indicating a degree of variation at a first time point in the value for the particular attribute of the disk; and
adding the attribute variation value to a second data item of the operation data corresponding to the first time point.