Data management in RAID environment

• US 9,952,950 B1
• Filed: 06/16/2015
• Issued: 04/24/2018
• Est. Priority Date: 09/08/2014
• Status: Active Grant

• Current Assignee: JP Morgan Chase Bank, N.A. (JP Morgan Chase & Co)
• Original Assignee: Western Digital Technologies Incorporated (Western Digital Corporation)
• Inventors: Spongr, John
• Primary Examiner(s): DOAN, DUC T
• Application Number:
  US14/741,157
• Time in Patent Office: 1,043 Days
• Field of Search: None
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 11/1451 by selection of backup contents
• CPC Class Codes: G06F 11/2058 using more than 2 mirrored copies
• CPC Class Codes: G06F 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data

Abstract

A data storage system includes a plurality of Data Storage Devices (DSDs) that each includes at least one disk for storing data. Data is received for storage in a data stripe across the plurality of DSDs with each DSD of the plurality of DSDs storing a different portion of the data stripe. A strip size is determined for each portion of the data stripe based on a radial location where the portion of the data stripe will be stored on the at least one disk of its respective DSD. The data stripe is stored across the plurality of DSDs with each portion of the data stripe arranged in a radial location on the at least one disk of its respective DSD so that, when the data stripe is accessed, the radial locations of access in the plurality of DSDs progress in the same radial direction.

Claims

1. A data storage system, comprising:

a plurality of Data Storage Devices (DSDs) having a total storage capacity available for storing user data among the plurality of DSDs with each DSD of the plurality of DSDs having a DSD storage capacity that contributes to the total storage capacity of the plurality of DSDs; and

a controller configured to;

in response to one of the plurality of DSDs being replaced by a replacement DSD having a DSD storage capacity greater than the DSD storage capacity of the replaced DSD;

assign a new strip size to each DSD of the plurality of DSDs including the replacement DSD based at least in part on a portion of a new total storage capacity for the plurality of DSDs that is contributed by the DSD storage capacity of the respective DSD;

receive data for storage in a new data stripe; and

store the new data stripe across the plurality of DSDs including the replacement DSD such that each DSD stores a different portion of the new data stripe having the new strip size assigned to the DSD.

2. The data storage system of claim 1, further comprising at least one mirroring DSD for storing a copy of data stored on a respective mirrored DSD of the plurality of DSDs, wherein the controller is further configured to store a copy of a portion of a data stripe on a mirroring DSD of the least one mirroring DSD such that the copy of the portion is stored at approximately the same radial location where the portion is stored on a disk of the mirrored DSD of the plurality of DSDs.

3. The data storage system of claim 2, wherein the controller is further configured to:

send a first read command to access a portion of the data stripe from a mirrored DSD of the plurality of DSDs;

send a second read command to a mirroring DSD of the at least one mirroring DSD storing a copy of the portion stored on the mirrored DSD;

initially receive one of the portion from the mirrored DSD or the copy of the portion from the mirroring DSD;

abort the first read command if the copy of the portion is initially received from the mirroring DSD; and

abort the second read command if the portion is initially received from the mirrored DSD.

4. The data storage system of claim 2, wherein the controller is further configured to:

receive a request from a host to store data in the data storage system;

send a first set of write commands to the plurality of DSDs to store an original data stripe for the data across the plurality of DSDs;

send a second set of write commands to a set of mirroring DSDs to store a copy of the original data stripe across the set of mirroring DSDs, wherein each mirroring DSD in the set of mirroring DSDs is paired with a corresponding DSD of the plurality of DSDs to store a copy of data stored in the corresponding DSD;

receive indications from the plurality of DSDs and the set of mirroring DSDs indicating the storage of a portion of the original data stripe or a portion of the copy of the original data stripe; and

send a completion notification to the host indicating that the data has been stored in the data storage system upon initially receiving indications for each pair of mirroring DSDs and corresponding DSDs of the plurality of DSDs.

5. The data storage system of claim 1, wherein the controller includes a memory, and wherein the controller is further configured to store at least one address in the memory for each portion of the new data stripe to identify the portion and to associate the portion with the DSD storing the portion.

6. The data storage system of claim 1, wherein the plurality of DSDs stores a plurality of data stripes across the plurality of DSDs and the controller includes a memory, and wherein the controller is further configured to store a starting address in the memory for each data stripe to indicate a beginning location of the data stripe with the starting address for each data stripe after a first data stripe being offset from a previous starting address for a previous data stripe by a predetermined address offset based on the number of data stripes in the plurality of data stripes.

7. The data storage system of claim 1, wherein the data stripe includes a stream of video data or logging data.

8. The data storage system of claim 1, wherein the controller is further configured to determine the plurality of DSDs from a larger set of DSDs by:

pairing each of one or more DSDs of the larger set with another DSD of the larger set that has the closest DSD storage capacity available for storing user data; and

for each pair of DSDs, setting a first DSD of the pair as a DSD of the plurality of DSDs and setting the other DSD of the pair as a mirroring DSD for storing a copy of user data stored on the first DSD.

9. A method of operating a data storage system including a plurality of Data Storage Devices (DSDs) having a total storage capacity available for storing user data among the plurality of DSDs with each DSD of the plurality of DSDs having a DSD storage capacity that contributes to the total storage capacity of the plurality of DSDs, the method comprising:

in response to one of the plurality of DSDs being replaced by a replacement DSD having a DSD storage capacity greater than the DSD storage capacity of the replaced DSD;

assigning a new strip size to each DSD of the plurality of DSDs including the replacement DSD based at least in part on a portion of a new total storage capacity for the plurality of DSDs that is contributed by the DSD storage capacity of the respective DSD;

receiving data for storage in a new data stripe; and

storing the new data stripe across the plurality of DSDs including the replacement DSD such that each DSD stores a different portion of the new data stripe having the new strip size assigned to the DSD.

10. The method of claim 9, wherein the data storage system further includes at least one mirroring DSD for storing a copy of data stored on a respective mirrored DSD of the plurality of DSDs, and wherein the method further comprises storing a copy of a portion of a data stripe on a mirroring DSD of the least one mirroring DSD such that the copy of the portion is stored at approximately the same radial location where the portion is stored on a disk of the mirrored DSD of the plurality of DSDs.

11. The method of claim 10, further comprising:

sending a first read command to access a portion of the data stripe from a mirrored DSD of the plurality of DSDs;

sending a second read command to a mirroring DSD of the at least one mirroring DSD storing a copy of the portion stored on the mirrored DSD;

initially receiving one of the portion from the mirrored DSD or the copy of the portion from the mirroring DSD;

aborting the first read command if the copy of the portion is initially received from the mirroring DSD; and

aborting the second read command if the portion is initially received from the mirrored DSD.

12. The method of claim 10, further comprising:

receiving a request from a host to store data in the data storage system;

sending a first set of write commands to the plurality of DSDs to store an original data stripe for the data across the plurality of DSDs;

sending a second set of write commands to a set of mirroring DSDs to store a copy of the original data stripe across the set of mirroring DSDs, wherein each mirroring DSD in the set of mirroring DSDs is paired with a corresponding DSD of the plurality of DSDs to store a copy of data stored in the corresponding DSD;

receiving indications from the plurality of DSDs and the set of mirroring DSDs indicating the storage of a portion of the original data stripe or a portion of the copy of the original data stripe; and

sending a completion notification to the host indicating that the data has been stored in the data storage system upon initially receiving indications for each pair of mirroring DSDs and corresponding DSDs of the plurality of DSDs.

13. The method of claim 9, further comprising storing at least one address in a memory for each portion of the new data stripe to identify the portion and to associate the portion with the DSD storing the portion.

14. The method of claim 9, wherein the plurality of DSDs stores a plurality of data stripes across the plurality of DSDs and the method further comprises storing a starting address in a memory for each data stripe to indicate a beginning location of the data stripe with the starting address for each data stripe after a first data stripe being offset from a previous starting address for a previous data stripe by a predetermined address offset based on the number of data stripes in the plurality of data stripes.

15. The method of claim 9, wherein the data stripe includes a stream of video data or logging data.

16. The method of claim 9, wherein the method further comprises determining the plurality of DSDs from a larger set of DSDs by:

pairing each of one or more DSDs of the larger set with another DSD of the larger set that has the closest DSD storage capacity available for storing user data; and

for each pair of DSDs, setting a first DSD of the pair as a DSD of the plurality of DSDs and setting the other DSD of the pair as a mirroring DSD for storing a copy of user data stored on the first DSD.

17. A controller for a data storage system including a plurality of Data Storage Devices (DSDs) having a total storage capacity available for storing user data among the plurality of DSDs with each DSD of the plurality of DSDs having a DSD storage capacity that contributes to the total storage capacity of the plurality of DSDs, the controller comprising:

an interface for communicating with the plurality of DSDs; and

a processor configured to;

in response to one of the plurality of DSDs being replaced by a replacement DSD having a DSD storage capacity greater than the DSD storage capacity of the replaced DSD;

assign a new strip size to each DSD of the plurality of DSDs including the replacement DSD based at least in part on a portion of a new total storage capacity for the plurality of DSDs that is contributed by the DSD storage capacity of the respective DSD;

receive data for storage in a new data stripe across the plurality of DSDs; and

store the new data stripe across the plurality of DSDs including the replacement DSD such that each DSD stores a different portion of the new data stripe having the new strip size assigned to the DSD.

18. A data storage system, comprising:

a plurality of Data Storage Devices (DSDs) having a total storage capacity available for storing user data among the plurality of DSDs with each DSD of the plurality of DSDs having a DSD storage capacity that contributes to the total storage capacity of the plurality of DSDs; and

a controller configured to;

store a plurality of data stripes across the plurality of DSDs; and

in response to one of the plurality of DSDs being replaced by a replacement DSD having a DSD storage capacity less than the DSD storage capacity of the replaced DSD;

determine a difference between the DSD storage capacity of the replaced DSD and the DSD storage capacity of the replacement DSD, wherein the total storage capacity is reduced to a total replacement storage capacity for storing user data among the plurality of DSDs other than the replaced DSD and the replacement DSD;

identify incomplete data stripes of the plurality of data stripes that would be incomplete due to the smaller DSD storage capacity of the replacement DSD;

assign a replacement strip size for the incomplete data stripes to each DSD of the plurality of DSDs other than the replaced DSD and the replacement DSD, wherein the replacement strip size is based at least in part on a portion of the total replacement storage capacity that is contributed by the DSD storage capacity of the respective DSD; and

store the incomplete data stripes across the plurality of DSDs other than the replaced DSD and the replacement DSD such that for each incomplete data stripe, each DSD stores a different replacement portion of the incomplete data stripe having the replacement strip size assigned to the DSD.

19. The data storage system of claim 18, wherein in identifying incomplete data stripes of the plurality of data stripes, the controller is further configured to determine the number of incomplete data stripes by dividing the difference between the DSD storage capacity of the replaced DSD and the DSD storage capacity of the replacement DSD by a data size for each of the plurality of data stripes.

20. A method of operating a data storage system including a plurality of Data Storage Devices (DSDs) having a total storage capacity available for storing user data among the plurality of DSDs with each DSD of the plurality of DSDs having a DSD storage capacity that contributes to the total storage capacity of the plurality of DSDs, the method comprising:

storing a plurality of data stripes across the plurality of DSDs; and

in response to one of the plurality of DSDs being replaced by a replacement DSD having a DSD storage capacity less than the DSD storage capacity of the replaced DSD;

determining a difference between the DSD storage capacity of the replaced DSD and the DSD storage capacity of the replacement DSD, wherein the total storage capacity is reduced to a total replacement storage capacity for storing user data among the plurality of DSDs other than the replaced DSD and the replacement DSD;

identifying incomplete data stripes of the plurality of data stripes that would be incomplete due to the smaller DSD storage capacity of the replacement DSD;

assigning a replacement strip size for the incomplete data stripes to each DSD of the plurality of DSDs other than the replaced DSD and the replacement DSD, wherein the replacement strip size is based at least in part on a portion of the total replacement storage capacity that is contributed by the DSD storage capacity of the respective DSD; and

storing the incomplete data stripes across the plurality of DSDs other than the replaced DSD and the replacement DSD such that for each incomplete data stripe, each DSD stores a different replacement portion of the incomplete data stripe having the replacement strip size assigned to the DSD.

21. The method of claim 20, wherein in identifying incomplete data stripes of the plurality of data stripes, the controller is further configured to determine the number of incomplete data stripes by dividing the difference between the DSD storage capacity of the replaced DSD and the DSD storage capacity of the replacement DSD by a data size for each of the plurality of data stripes.

22. A controller for a data storage system including a plurality of Data Storage Devices (DSDs) having a total storage capacity available for storing user data among the plurality of DSDs with each DSD of the plurality of DSDs having a DSD storage capacity that contributes to the total storage capacity of the plurality of DSDs, the controller comprising:

an interface for communicating with the plurality of DSDs; and

a processor configured to;

store a plurality of data stripes across the plurality of DSDs; and

in response to one of the plurality of DSDs being replaced by a replacement DSD having a DSD storage capacity less than the DSD storage capacity of the replaced DSD;

determine a difference between the DSD storage capacity of the replaced DSD and the DSD storage capacity of the replacement DSD, wherein the total storage capacity is reduced to a total replacement storage capacity for storing user data among the plurality of DSDs other than the replaced DSD and the replacement DSD;

identify incomplete data stripes of the plurality of data stripes that would be incomplete due to the smaller DSD storage capacity of the replacement DSD;

assign a replacement strip size for the incomplete data stripes to each DSD of the plurality of DSDs other than the replaced DSD and the replacement DSD, wherein the replacement strip size is based at least in part on a portion of the total replacement storage capacity that is contributed by the DSD storage capacity of the respective DSD; and

store the incomplete data stripes across the plurality of DSDs other than the replaced DSD and the replacement DSD such that for each incomplete data stripe, each DSD stores a different replacement portion of the incomplete data stripe having the replacement strip size assigned to the DSD.

23. The controller of claim 22, wherein in identifying incomplete data stripes of the plurality of data stripes, the processor is further configured to determine the number of incomplete data stripes by dividing the difference between the DSD storage capacity of the replaced DSD and the DSD storage capacity of the replacement DSD by a data size for each of the plurality of data stripes.