Monitoring health condition of a hard disk
Various embodiments of the present disclosure provide a method and apparatus for monitoring health condition of a hard disk by obtaining full-dimensional characteristics associated with the hard disk, wherein the full-dimension characteristics comprise at least two of: hard disk performance information, data integrity information, input/output I/O correctness information, and a hard disk Self-Monitoring Analysis and Reporting Technology S.M.A.R.T. report, and determining the health condition of the hard disk based on the full-dimension characteristics.
1. A method for monitoring health condition of a hard disk in a storage system, the method comprising:
obtaining characteristics associated with a hard disk from a host bus adapter (HBA) configured to transport data between the hard disk and a bus, wherein the characteristics comprise at least two of:
hard disk performance information, data integrity information, input/output I/O correctness information, and a hard disk Self-Monitoring Analysis and Reporting Technology S.M.A.R.T. report, wherein the characteristics are obtained from the HBA to avoid obtaining the hard disk performance information affected by a higher layer of a software stack operating on the storage system;
wherein the I/O correctness information refers to correctness or error of an I/O path which is taken in the course of reading data from or writing data to the hard disk, wherein the I/O correctness information is enabled to be collected from the small computer system interface SCSI protocol layer and the HBA; and
determining a health condition of the hard disk based on the characteristics to avoid an erroneous assessment of the health condition of the hard disk based on the S.M.A.R.T report.
2. The method according to claim 1 , wherein the hard disk performance information includes at least one of:
average I/O latency, I/O throughput, an average I/O size and a proportion of an I/O pattern, wherein the I/O pattern includes a sequential read/write pattern and a random read/write pattern.
3. The method according to claim 1 , wherein the data integrity information is obtained from a Redundant Array of Independent Disks (RAID) corresponding to the hard disk, and the data integrity information obtained from an upper layer system of the RAID.
4. The method according to claim 1 , wherein the data integrity information includes information of data integrity verified by the RAID.
5. The method according to claim 1 , wherein the I/O correctness information is obtained from at least one of an HBA drive program and a small computer system interface SCSI protocol layer corresponding to the hard disk, and wherein the I/O correctness information includes at least one of:
HBA hardware/firmware-related errors, link-related errors and hard disk medium-related errors.
6. The method according to claim 1 , wherein the hard disk S.M.A.R.T. report is obtained from the hard disk wherein the hard disk S.M.A.R.T. report is associated with the type of the hard disk.
7. The method according to claim 1 , wherein before obtaining the characteristics associated with the hard disk, the method comprising:
pre-collecting the characteristics periodically or according to a trigger by an event; and
storing the characteristics.
8. The method according to claim 7 , wherein determining the health condition of the hard disk based on the characteristics comprises at least one of:
determining the health condition of the hard disk periodically based on the characteristics; OR
determining the health condition of the hard disk based on the characteristics according the trigger by an event.
9. An apparatus for monitoring health condition of a hard disk in a storage system, comprising:
an analysis unit configured for obtaining characteristics associated with a hard disk from a host bus adapter (HBA) configured to transport data between the hard disk and a bus, wherein the characteristics comprise at least two of:
hard disk performance information, data integrity information, input/output I/O correctness information, and a hard disk Self-Monitoring Analysis and Reporting Technology S.M.A.R.T. report, wherein the characteristics are obtained from the HBA to avoid obtaining the hard disk performance information affected by a higher layer of a software stack operating on the storage system;
wherein the I/O correctness information refers to correctness or error of an I/O path which is taken in the course of reading data from or writing data to the hard disk, wherein the I/O correctness information is enabled to be collected from the small computer system interface SCSI protocol layer and the HBA; and
determining the health condition of the hard disk based on the characteristics to avoid an erroneous assessment of the health condition of the hard disk based on the S.M.A.R.T report.
10. The apparatus according to claim 9 , wherein the hard disk performance information includes at least one of:
average I/O latency, I/O throughput, an average I/O size and a proportion of an I/O pattern, wherein the I/O pattern includes a sequential read/write pattern and a random read/write pattern.
11. The apparatus according to claim 9 , wherein the data integrity information is obtained from a Redundant Array of Independent Disks (RAID) corresponding to the hard disk, and wherein the data integrity information is further obtained from an upper layer system of the RAID.
12. The apparatus according to claim 9 , wherein the data integrity information includes statistics results of data integrity verified by the RAID.
13. The apparatus according to claim 9 , wherein the I/O correctness information is obtained from at least one of a HBA drive program and a small computer system interface SCSI protocol layer corresponding to the hard disk, and wherein the I/O correctness information includes at least one of: HBA hardware/firmware-related errors, link-related errors and hard disk medium-related errors.
14. The apparatus according to claim 9 , wherein the hard disk S.M.A.R.T. report is obtained from the hard disk, and wherein the hard disk S.M.A.R.T. report is associated with the type of the hard disk.
15. The apparatus according to claim 9 , further configured for
pre-collecting the characteristics periodically or according to a trigger by an event; and
storing the characteristics.
16. The apparatus according to claim 15 , comprises at least one of determining the health condition of the hard disk periodically based on the characteristics; OR determining the health condition of the hard disk based on the characteristics according to the trigger by an event.
17. A computer program product for facilitating management of resources, the computer program product being tangibly stored on a non-transient computer-readable medium and comprising machine executable instructions which, when executed, causing the machine to perform a method for monitoring health condition of a hard disk in a storage system, comprising:
obtaining characteristics associated with a hard disk from a host bus adapter (HBA) configured to transport data between the hard disk and a bus, wherein the characteristics comprise at least two of:
hard disk performance information, data integrity information, input/output I/O correctness information, and a hard disk Self-Monitoring Analysis and Reporting Technology S.M.A.R.T. report, wherein the characteristics are obtained from the HBA to avoid obtaining the hard disk performance information affected by a higher layer of a software stack operating on the storage system, and
wherein the hard disk performance information is obtained from a host bus adapter (HBA) drive program corresponding to the hard disk, and wherein the hard disk performance information includes at least one of: average I/O latency, I/O throughput, an average I/O size and a proportion of an I/O pattern, wherein the I/O pattern includes a sequential read/write pattern and a random read/write pattern;
wherein I/O correctness information is obtained from at least one of the HBA drive program and a small computer system interface SCSI protocol layer corresponding to the hard disk, and wherein the I/O correctness information includes at least one of: HBA hardware/firmware-related errors, link-related errors and hard disk medium-related errors, wherein the I/O correctness information is enabled to be collected from the small computer system interface SCSI protocol layer and the HBA;
wherein the I/O correctness information refers to correctness or error of an I/O path which is taken in the course of reading data from or writing data to the hard disk;
wherein the hard disk S.M.A.R.T. report is obtained from the hard disk wherein the hard disk S.M.A.R.T. report is associated with the type of the hard disk; and
determining a health condition of the hard disk based on the characteristics to avoid an erroneous assessment of the health condition of the hard disk based on the S.M.A.R.T report.
18. The computer program product to claim 17 , wherein the data integrity information is obtained from a Redundant Array of Independent Disks (RAID) corresponding to the hard disk, and the data integrity information obtained from an upper layer system of the RAID.
19. The computer program product of claim 17 , wherein the machine executable instructions which, when executed, further causing the machine to perform a method for monitoring health condition of a hard disk:
wherein before obtaining the characteristics associated with the hard disk:
pre-collecting the characteristics periodically or according to a trigger by an event; and
storing the characteristics.
20. The computer program product of claim 19 , wherein determining the health condition of the hard disk based on the characteristics comprises at least one of:
determining the health condition of the hard disk periodically based on the characteristics; OR
determining the health condition of the hard disk based on the characteristics according the trigger by an event.