Disk scrubbing system

• US 5,632,012 A
• Filed: 03/06/1996
• Issued: 05/20/1997
• Est. Priority Date: 11/24/1993
• Status: Expired due to Term

• Current Assignee: Storage Technologies Corporation (Oracle Corporation)
• Original Assignee: Storage Technologies Corporation (Oracle Corporation)
• Inventors: Rudeseal, George A., Belsan, Jay S., Permut, Alan R.
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): Elmore, Stephen C.
• Application Number:
  US08/611,501
• Time in Patent Office: 440 Days
• Field of Search: 395/182.04, 395/182.01, 371/40.1, 371/40.2, 371/10.2, 364/266.3, 364/267.91, 364/266.5, 364/970
• US Class Current: 714/6.12
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 2211/1088 Scrubbing in RAID systems with parity
• CPC Class Codes: G11B 20/18 Error detection or correction; Testing , e.g. of drop-outs
• CPC Class Codes: G11B 20/1833 by adding special lists or symbols to the coded information G11B20/1806, G11B20/1866 take precedence

Abstract

One or more background processes that references data storage volumes and cylinders that are to be scrubbed to ensure the integrity of the data stored thereon. A priority scrubbing process is also available to note data storage volumes that have recently had data written thereon by the host processor and which require a more timely review of the data than less frequently used volumes.

Claims

1. A data storage subsystem that receives data from at least one connected data processor and stores said data on a plurality of disk drives divided into a plurality of logical partitions, said logical partitions comprising dedicated partitions currently storing data and free partitions available to store data, said data storage subsystem comprising:

a memory controller independent of said plurality of disk drives comprising means for identifying said dedicated and said free partitions;

means for reading said memory controller to determine the identity of said dedicated partitions; and

means responsive to said reading of said memory controller for periodically verifying the integrity of data currently stored in each of said identified dedicated partitions, said means for periodically verifying comprising;

means for reading data in each of said identified dedicated partitions;

means responsive to said reading of data from each said identified dedicated partition for generating error check information from said read data;

means responsive to said generation of said error check information for detecting errors in data in each said identified dedicated partition; and

means responsive to said detection of errors for correcting said data containing errors.

2. The data storage subsystem of claim 1 further comprising error check data generated from data written on said disk drives and stored on said disk drives, said means for detecting comprises:

means for comparing said generated error check information with said error check data stored on said disk drives, a mismatch between said generated error check information and said error check data being indicative of an error in said read data.

3. The data storage subsystem of claim 1 wherein said means for periodically verifying verifies all said identified dedicated partitions on a periodic cycle basis, said means for identifying comprises:

means for storing data that indicates a mapping between each said logical partition and a physical segment of a corresponding disk drive; and

means for recording at least one flag bit, corresponding to a one of said logical partitions and indicative of whether said one logical partition is verifiable in a presently active periodic cycle.

4. The data storage system of claim 1 further comprising:

means for noting an identification of each of said dedicated partitions in which data has been written since the data integrity of said dedicated partitions was last verified; and

means for priority verifying integrity of data written in each of said noted dedicated partitions, wherein said means for priority verifying includes means for sequencing through all data in each of said noted dedicated partitions to verify the integrity of data written in each said noted dedicated partition.

5. The data storage subsystem of claim 1 further comprising:

a plurality of disk drives for storing data, a number of said disk drives being configured into at least one redundancy group, each said at least one redundancy group including N+M of said plurality of disk drives, where N and M are both positive integers with N greater than 1 and M at least equal to 1;

means for storing a stream of data received from said at least one connected data processor on said N disk drives in a selected redundancy group; and

means, responsive to storing of said data on all N disk drives in said selected redundancy group, for writing M segments of redundancy data, computed from said data stored on said N disk drives in said selected redundancy group, on to said M disk drives of said selected redundancy group.

6. The data storage subsystem of claim 5 wherein said means for periodically verifying verifies data written on to said N disk drives as well as data written on to said M disk drives in a redundancy group.

7. The data storage subsystem of claim 1 wherein said means for periodically verifying further comprises a plurality of operationally independent concurrently operational processes.

8. The data storage subsystem of claim 7 wherein said means for periodically verifying further comprises:

means for adjusting the number of said concurrently operational processes as a function of a number of said plurality of disk drives installed in said data storage subsystem.

9. The data storage subsystem of claim 1 further comprising:

means for reading said memory controller to determine the identity of said free partitions; and

means responsive to said reading of said memory controller for periodically verifying the integrity of administrative data stored in each of said identified free partitions, said means for periodically verifying further comprising;

means for identifying administrative data stored in each of said identified free partitions;

means responsive to said identification of administrative data for reading said administrative data;

means responsive to said reading of said administrative data for generating error check information for said administrative data;

means responsive to said generation of error check information for detecting errors in administrative data for each identified free partition using said error check information; and

means responsive to said detection of errors for correcting errors contained in said administrative data.

10. A data storage subsystem that receives data from at least one connected data processor and stores said data on a plurality of disk drives divided into a plurality of logical partitions, each of said logical partitions comprising at least two disk drives, said data storage subsystem comprising:

said logical partitions further comprising dedicated partitions currently storing data and free partitions available for storing data;

a memory controller independent of said plurality of disk drives comprising means for identifying said dedicated partitions and said free partitions;

means for locating memory space available to store data in said logical partitions;

means for receiving data from said at least one connected data processor;

means responsive to said receiving of data for selecting a memory space in said logical partitions capable of storing the received data;

means for writing received data into said selected memory space;

means responsive to said writing of data for updating said identifying means to indicate that said free partitions containing said selected memory space have changed to currently dedicated partitions;

means for reading said memory controller to determine the identity of said dedicated partitions; and

means responsive to reading of said memory controller for periodically verifying the integrity of data currently stored in each of said identified dedicated partitions, said means for periodically verifying comprising;

means for reading data in each of said identified dedicated partitions;

means responsive to said reading of data from each said identified dedicated partition for generating error check information from said read data;

means responsive to said generation of said error check information for detecting errors in data in each said identified dedicated partition; and

means responsive to said detection of errors for

correcting said data containing errors.

11. The data storage subsystem of claim 10 further comprising:

means for recording at least one flag bit corresponding to each of said dedicated partitions in which data has been written since said each dedicated partition was last verified;

means for noting a state of said at least one flag bit of each of said dedicated partitions indicating that data has been written in a presently active periodic cycle; and

means for priority verifying integrity of data written in said noted dedicated partitions, wherein said means for priority verifying includes means for sequencing through all data in each of said noted dedicated partitions.

12. The data storage subsystem of claim 11 wherein said priority verifying means further comprises:

means for altering said state of said at least one flag bit when a corresponding dedicated partition has been verified to indicate the data written in said corresponding dedicated partition has been verified in said presently active periodic cycle.

13. The data storage subsystem of claim 11 wherein said means for priority verifying operates concurrently with and independently of said means for periodically verifying.

14. The data storage subsystem of claim 10 further comprising:

means for reading said memory controller to determine said identity of said free partitions;

means responsive to said reading of said memory controller for periodically verifying the integrity of administrative data stored in each of said identified free partitions, said means for periodically verifying further comprising;

means for identifying administrative data stored in each of said free partitions;

means responsive to said identification of administrative data for reading said administrative data;

means responsive to said reading of said administrative data for generating error check information for said administrative data;

means responsive to said generation of error check information for detecting errors in administrative data of each said identified free partition using said error check information; and

means responsive to said detection of errors for correcting errors contained in said administrative data.

15. A method for correcting data errors in a data storage subsystem that receives data from at least one connected data processor and stores said data on a plurality of disk drives divided into a plurality of logical partitions, said logical partitions comprising dedicated partitions currently storing data and free partitions available for storing data, the method comprising the steps of:

generating information identifying said dedicated partitions and said free partitions;

storing said generated information in a memory controller independent of said plurality of disk drives;

reading said memory controller to determine the identity of said dedicated partitions; and

periodically verifying the integrity of the data stored in each of said identified dedicated partitions comprising the steps of;

reading data stored in each of said identified dedicated partitions;

generating error check information from said read data;

detecting errors in data stored in each of said dedicated partitions using said generated error check information; and

correcting errors in said data stored in each said dedicated partition responsive to said detection of errors.

16. The method of claim 15 wherein each of said disk drives stores error check data generated from data written on said disk drives, said step of detecting errors comprises:

comparing said generated error check information with said error check data stored on said disk drives, a mismatch between said generated error check information and said error check data being indicative of an error in said read data.

17. The method of claim 15 wherein said step of periodically verifying verifies all said identified dedicated partitions on a periodic cycle basis, said step of identifying comprises:

storing data that indicates a mapping between a logical partition and a physical segment of a corresponding disk drive; and

recording at least one flag bit corresponding to a one of said dedicated partitions and indicative of whether said one dedicated partition is verifiable in a presently active periodic cycle.

18. The method of claim 15 further comprising the steps of:

noting an identification of each of said dedicated partitions in which data has been written since the data integrity of each said dedicated partition was last verified; and

priority verifying integrity of data written in each of said noted dedicated partitions by sequencing through all data in each of said noted dedicated partitions.

19. The method of claim 15, wherein said data storage subsystem includes a plurality of disk drives for storing data, a number of said disk drives being configured into at least one redundancy group, each said at least one redundancy group including N+M of said plurality of disk drives, where N and M are both positive integers with N greater than 1 and M at least equal to 1, said method further comprising the steps of:

storing a stream of data received from said at least one data processor on said N disk drives in a selected redundancy group;

writing M segments of redundancy data, said M segments generated from said data stored on said N disk drives in said selected redundancy group, on to said M disk drives of said selected redundancy group.

20. The method of claim 19 wherein said step of periodically verifying verifies data written on to said N disk drives as well as said M disk drives in a redundancy group.

21. The method of claim 15 wherein said step of periodically verifying comprises a plurality of operationally independent concurrently operational processes.

22. The method of claim 21 wherein said step of periodically verifying further comprises the step of:

adjusting the number of said concurrently operational processes as a function of a number of said plurality of disk drives installed in said data storage subsystem.

23. The method of claim 15 further comprising the steps of:

reading said memory controller to determine the identity of said free partitions; and

periodically verifying administrative data contained in each of said free partitions responsive to said identifying of said free partitions, said step of periodically verifying further comprising the steps of;

identifying administrative data contained in each said identified free partition;

reading said administrative data stored in each identified free partition responsive to said identification of said administrative data;

generating error check information for said administrative data responsive to said reading of said administrative data;

detecting errors contained in said administrative data using said error check information; and

correcting errors in said administrative data responsive to said detection of errors.

24. A method for correcting data errors in a data storage subsystem that receives data from at least one connected data processor and stores said data on a plurality of disk drives divided into a plurality of logical partitions, said logical partitions comprising dedicated partitions currently storing data and free partitions available to store data, said method comprising the steps of:

generating information identifying said dedicated partitions and said free partitions;

storing said generated information in a memory controller independent of said plurality of disk drives;

locating available memory space in said logical partitions;

receiving data from said at least one connected data processor;

selecting memory space from said located available memory space capable of storing said received data;

writing said received data into said selected memory space;

updating said generated information to indicate that said identified free partitions in which said received data was written have changed to currently dedicated partitions;

reading said memory controller to determine the identity of said dedicated partitions; and

periodically verifying the integrity of data stored on each of said identified dedicated partitions, said periodically verifying step comprising the steps of;

reading data stored in each of said identified dedicated partitions;

generating error check information from said read data;

detecting errors in data stored in each of said dedicated partitions using said generated error check information; and

correcting errors in data stored in each said dedicated partition responsive to said detection of errors.

25. The method of claim 24 further comprising the steps of:

recording at least one flag bit corresponding to each of said dedicated partitions in which data has been written since said each dedicated partition was last verified;

noting a state of said at least one flag bit of each of said dedicated partitions indicating that data has been written in a presently active periodic cycle; and

priority verifying integrity of data written in each of said noted dedicated partitions by sequencing through all data in each of said noted dedicated partitions.

26. The method of claim 24 wherein said step of priority verifying further comprises:

altering said state of said at least one flag bit when a corresponding dedicated partition has been verified to indicate the data written in said corresponding dedicated partition has been verified in said presently active periodic cycle.

27. The method of claim 25 wherein said step of priority verifying operates concurrently with and independent of said step of periodically verifying.

28. The method of claim 24 further comprising the steps of:

reading said memory controller to determine the identity of said free partitions; and

periodically verifying administrative data contained in each of said free partitions responsive to said identifying of said free partitions, said step of periodically verifying further comprising the steps of;

identifying said administrative data contained in each said identified free partition;

reading said administrative data stored in each identified free partition responsive to said identification of said administrative data;

generating error check information from said administrative data responsive to said reading of said administrative data;

detecting errors contained in said administrative data using said error check information; and

correcting errors in said administrative data responsive to said detection of errors.

29. A data storage subsystem that receives data for at least one connected data processor and stores said data on a plurality of disk drives divided into a plurality of logical partitions, said logical partitions comprising dedicated partitions currently storing data and free partitions available for storing data, said data storage subsystem comprising:

a virtual track directory for mapping said plurality of logical partitions to corresponding physical segments on said plurality of disk drives;

a plurality of flag bits each corresponding to one of said logical partitions, stored in a memory controller independent of said plurality of disk drives, for identifying said dedicated partitions and free partitions;

means for reading each said flag bit to determine the identity of said dedicated partitions;

means responsive to said reading of each said flag bit for periodically verifying the integrity of data stored in each of said identified dedicated partitions, said verifying means comprising;

means for reading data in each of said identified dedicated partitions;

means responsive to said reading of data from each said identified dedicated partition for generating error check information from said read data;

means responsive to said generation of said error check information for detecting errors in data stored in each said identified dedicated partition; and

means responsive to said detection of errors for correcting said data containing errors.

30. The data storage subsystem of claim 29 further comprising error check data stored on said disk drives and generated from data written on said disk drives, and said detecting means comprises:

means for comparing said generated error check information with said error check data stored on said disk drives, a mismatch between said generated error check information and said error check data being indicative of an error in said read data.

31. The data storage subsystem of claim 29 further comprising:

means for noting an identification of each of said dedicated partitions in which data has been written since the data integrity of said each dedicated partition was last verified; and

means for priority verifying integrity of data written in each of said noted dedicated partitions, wherein said means for priority verifying includes means for sequencing through all data in each of said noted dedicated partitions.

32. The data storage subsystem of claim 29 further comprising:

a plurality of disk drives for storing data, a number of said disk drives being configured into at least one redundancy group, each said at least one redundancy group including N+M of said plurality of disk drives, where N and M are both positive integers with N greater than 1 and M at least equal to 1;

means for storing a stream of data received from said at least one data processor on said N disk drives in a selected redundancy group;

means, responsive to said storing of said data on all N disk drives in said selected redundancy group, for writing M segments of redundancy data, computed from said data stored on said N disk drives in said selected redundancy group, on to said M disk drives of said selected redundancy group.

33. The data storage subsystem of claim 32 wherein said means for periodically verifying verifies data written on to said N disk drives as well as data written on to said M disk drives in a redundancy group.

34. The data storage subsystem of claim 29 wherein said means for periodically verifying comprises a plurality of operationally independent concurrently operational processes.

35. The data storage subsystem of claim 34 wherein said means for periodically verifying further comprises:

means for adjusting the number of said concurrently operational processes as a function of a number of said plurality of disk drives installed in said data storage subsystem.

36. The data storage subsystem of claim 29 further comprising:

means for reading said flag bits to identify said free partitions;

means responsive to said reading of said flag bit for periodically verifying the integrity of administrative data stored in each of said identified free partitions, said means for periodically verifying further comprising;

means for identifying said administrative data stored in each of said identified free partitions;

means responsive to said identification of said administrative data for reading said administrative data;

means responsive to said reading of data for generating error check information from said administrative data;

means for detecting errors in said administrative data using said error check information; and

means responsive to said detection of errors for correcting errors contained in said administrative data.