MASS STORAGE SYSTEM AND METHOD OF OPERATING THEREOF

• US 20100287408A1
• Filed: 05/10/2010
• Published: 11/11/2010
• Est. Priority Date: 05/10/2009
• Status: Active Grant

• Current Assignee: Infinidat Ltd.
• Original Assignee: Xsignnet Ltd.
• Inventors: KOPYLOVITZ, Haim
• Granted Patent: US 8,495,295 B2
• Time in Patent Office: Days
• US Class Current: 714/6
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3485 for I/O devices
• CPC Class Codes: G06F 2201/815 Virtual
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: G06F 2201/875 Monitoring of systems including the internet
• CPC Class Codes: G06F 2201/885 Monitoring specific for caches
• CPC Class Codes: G06F 2211/1057 Parity-multiple bits-RAID6, i.e. RAID 6 implementations

Abstract

There are provided a mass storage system and a method of operating thereof. The method comprises: a) dividing the storage space into a first portion configured to be available to a client and a second portion configured to be unavailable to a client, thus giving rise, respectively to an available storage space and a spare storage space; b) distributing the available space and the spare space over the disk drives thus giving rise to available space and spare space allocated to each given disk drive; c) detecting underperformance of at least one disk drive among the plurality of disk drives; d) responsive to detecting underperformance, decreasing respective available space allocated to said at least one underperforming disk drive and respectively increasing spare space allocated to said at least one underperforming disk drive thus giving rise to a re-configured allocation of the available space; and e) transferring the stored data in accordance with said re-configured allocation of the available space.

Claims

1. A method of operating a mass storage system comprising a storage layer operatively coupled to a plurality of disk drives constituting a storage space, the system configured to enable direct or indirect serving a request directed to any accessible address in the storage space, the method comprising:

a. dividing the storage space into a first portion configured to be available to a client and a second portion configured to be unavailable to a client, thus giving rise, respectively to an available storage space and a spare storage space;

b. distributing the available space and the spare space over the disk drives thus giving rise to available space and spare space allocated to each given disk drive;

c. detecting underperformance of at least one disk drive among the plurality of disk drives;

d. responsive to detecting underperformance, decreasing respective available space allocated to said at least one underperforming disk drive and respectively increasing spare space allocated to said at least one underperforming disk drive thus giving rise to a re-configured allocation of the available space; and

e. transferring the stored data in accordance with said re-configured allocation of the available space.

2. The method of claim 1 further comprising distributing a remainder of the available space resulting from said re-configured allocation of the available space over at least some of due performing disks drives among the plurality of disk drives and respectively decreasing the spare space allocated to said drives.

3. The method of claim 2 wherein distributing the available space and the spare space over the disk drives is provided evenly.

4. The method of claim 3 wherein distributing said remainder of the available space over at least part of the due performing disks and respectively decreasing the spare space allocated to said drives are provided evenly.

5. The method of claim 1 wherein the amount of increased spare space allocated to said at least one underperforming disk drive is predefined as a certain percentage of respective allocated spare space.

6. The method of claim 1 wherein the amount of increased spare space allocated to said at least one underperforming disk drive is predefined as a function of at least one parameter selected from a group comprising:

a. a number of currently underperforming disk drives among the plurality of disk drives;

b. a number of underperforming disk drives among the plurality of disk drives detected during a certain period of time;

c. a rate of deviation of the detected level of underperformance from the current average performance of the plurality of the disk drives; and

d. a rate of deviation of the detected level of underperformance from the average performance of the plurality of the disk drives in the system over a certain period of time.

7. The method of claim 1 wherein the stored data portions are arranged in one or more RAID groups with the help of an allocation function and wherein transferring the stored data portions comprises:

a. re-calculating the values of the allocation function in accordance with said re-configured allocation of available space over the disk drives;

b. detecting members of RAID groups which need to be moved in accordance with the re-calculated values of the allocation function; and

c. moving data corresponding to said detected members of RAID groups in accordance with the re-calculated values of the allocation function

8. The method of claim 1 wherein detecting underperformance of at least one disk drive is provided with the help of SMART indication.

9. The method of claim 1 wherein underperformance is detected if an estimated value characterizing expected performance degradation matches a certain underperformance criteria.

10. The method of claim 9 wherein READ ERROR RATE and/or SEEK ERROR RATE are used as parameters indicative of expected performance degradation for SMART-enabled disk drives.

11. The method of claim 10 wherein a rate of variation of values of READ ERROR RATE and/or SEEK ERROR RATE is used as said estimated value characterizing expected performance degradation.

12. The method of claim 11 wherein said rate of variation is calculated in relation with a certain rate measuring threshold.

13. The method of claim 12 wherein READ ERROR RATE values and/or SEEK ERROR RATE values respective rate measuring thresholds used for estimating expected performance degradation substantially before achieving a threshold predefined for alerting an expected failure of SMART-enabled disk drive.

14. A mass storage system comprising:

a. a plurality of disk drives constituting a storage space, wherein;

i. the system configured to enable direct or indirect serving a request directed to any accessible address in the storage space;

ii. the storage space is divided into a first portion configured to be available to a client and a second portion configured to be unavailable to a client, thus giving rise, respectively to an available storage space and a spare storage space;

iii, the available space and the spare space are distributed over the disk drives thus giving rise to available space and spare space allocated to each given disk drive;

b. means for detecting underperformance of at least one disk drive among the plurality of disk drives;

c. a control layer operatively coupled to the plurality of disk drives and configured;

i. responsive to detecting underperformance, to decrease respective available space allocated to said at least one underperforming disk drive and respectively increase spare space allocated to said at least one underperforming disk drive thus giving rise to a re-configured allocation of the available space; and

ii. facilitate transferring the stored data in accordance with said re-configured allocation of the available space.

15. The system of claim 14 wherein the control layer is further configured to distribute a remainder of the available space resulting from said re-configured allocation of the available space over at least some of due performing disks drives among the plurality of disk drives and respectively decrease the spare space allocated to said drives.

16. The system of claim 15 wherein distributing the available space and the spare space over the disk drives, distributing said remainder of the available space over at least part of the due performing disks and respectively decreasing the spare space allocated to said drives are provided evenly

17. The system of claim 14 wherein the amount of increased spare space allocated to said at least one underperforming disk drive is predefined as a certain percentage of respective allocated spare space.

18. The system of claim 14 wherein the amount of increased spare space allocated to said at least one underperforming disk drive is predefined as a function of at least one parameter selected from a group comprising:

a. a number of currently underperforming disk drives among the plurality of disk drives;

b. a number of underperforming disk drives among the plurality of disk drives detected during a certain period of time;

c. a rate of deviation of the detected level of underperformance from the current average performance of the plurality of the disk drives; and

d. a rate of deviation of the detected level of underperformance from the average performance of the plurality of the disk drives in the system over a certain period of time.

19. The system of claim 14 wherein the stored data portions are arranged in one or more RAID groups with the help of an allocation function and wherein transferring the stored data portions comprises:

a. re-calculating the values of the allocation function in accordance with said re-configured allocation of available space over the disk drives;

b. detecting members of RAID groups which need to be moved in accordance with the re-calculated values of the allocation function; and

c. moving data corresponding to said detected members of RAID groups in accordance with the re-calculated values of the allocation function.

20. The system of claim 14 wherein said means for detecting underperformance of disk drives comprise one or more SMART controllers integrated with individual disk drives among said plurality of disk drives.

21. The system of claim 14 wherein at least part of said means for detecting underperformance of disk drives is comprised in said control layer.

22. The system of claim 14 further configured to estimate a value characterizing expected performance degradation of a given disk drive and to detect underperformance of said disk drive if said estimated value matches a certain underperformance criteria.

23. The system of claim 22 wherein READ ERROR RATE and/or SEEK ERROR RATE are used as parameters indicative of expected performance degradation for SMART-enabled disk drives.

24. The system of claim 23 wherein a rate of variation of values of READ ERROR RATE and/or SEEK ERROR RATE is used as said estimated value characterizing expected performance degradation.

25. The system of claim 24 wherein said rate of variation is calculated in relation with a certain rate measuring threshold.