Coordinated garbage collection for raid array of solid state disks

• US 8,713,268 B2
• Filed: 01/28/2011
• Issued: 04/29/2014
• Est. Priority Date: 08/05/2010
• Status: Active Grant

• Current Assignee: UT-Battelle LLC
• Original Assignee: UT-Battelle LLC
• Inventors: Dillow, David A., Kim, Youngjae, Oral, Hakki S., Shipman, Galen M., Wang, Feiyi
• Primary Examiner(s): Elmore, Reba I
• Application Number:
  US13/015,750
• Publication Number:
  US 20120036309A1
• Time in Patent Office: 1,187 Days
• Field of Search: 711/114, 711/152, 711/165
• US Class Current: 711/162
• CPC Class Codes: G06F 11/108 Parity data distribution in semiconductor storages, e.g. in SSD
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 12/0253 Garbage collection, i.e. reclamation of unreferenced memory
• CPC Class Codes: G06F 2211/1052 RAID padding, i.e. completing a redundancy group with dummy data
• CPC Class Codes: G06F 2212/7205 Cleaning, compaction, garbage collection, erase control
• CPC Class Codes: G06F 3/0652 Erasing, e.g. deleting, data cleaning, moving of data to a wastebasket
• CPC Class Codes: G06F 3/0679 Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays

Abstract

An optimized redundant array of solid state devices may include an array of one or more optimized solid-state devices and a controller coupled to the solid-state devices for managing the solid-state devices. The controller may be configured to globally coordinate the garbage collection activities of each of said optimized solid-state devices, for instance, to minimize the degraded performance time and increase the optimal performance time of the entire array of devices.

Claims

1. An optimized redundant array of solid state drives comprising:

an array of two or more optimized solid state drives; and

a controller coupled to the solid state drives for managing the solid state drives, said controller configured to globally coordinate the garbage collection activities of each of said optimized solid-state drives participating in the globally coordinated garbage collection.

2. The optimized redundant array of solid state drives of claim 1, wherein the garbage collection is coordinated by the controller in an active mode in which one or more of the optimized solid-state drives communicate to said controller a need to initiate a garbage collection activity, and said controller in response to receiving the communication, requests all of the solid-state drives in the array to initiate a garbage collection activity so that garbage collection cycles begin on all of the solid-state drives in the array at a scheduled time.

3. The optimized redundant array of solid state drives of claim 2, wherein said controller in response to receiving the communication, delays requesting all of the solid-state drives in the array until another communication for a need to initiate a garbage collection activity is received from another one or more of said optimized solid state drives.

4. The optimized redundant array of solid state drives of claim 2, wherein said controller in response to receiving the communication, delays requesting all of the solid-state drives in the array to begin the garbage collection cycles.

5. The optimized redundant array of solid state drives of claim 1, wherein the garbage collection is coordinated by the controller in a passive mode, in which the controller periodically collects information from at least one of said two or more solid state drives in the array to determine an appropriate time to initiate a global garbage collection cycle and initiates the global garbage collection cycle at said determined appropriate time on said two or more solid state drives in the array.

6. The optimized redundant array of solid state drives of claim 5, wherein the controller collects by pulling said ene two or more solid state drives for the information, or said one or more solid state drives pushing the information to the controller, or combinations thereof.

7. The optimized redundant array of solid state drives of claim 6, wherein the information includes internal fragmentation level, number of free erase blocks available, or ECC correctable error rate on reads, or combinations thereof.

8. The optimized redundant array of solid state drives of claim 1, wherein said controller initiate global garbage collection on said array by communicating to each of said solid state drives a specified time to begin a garbage collection activity.

9. The optimized redundant array of solid state drives of claim 1, wherein said controller identifies one or more points in input/output stream associated with the solid state drives that are expected to have idle periods and initiates a global garbage collection activity to be performed during the identified one or more points.

10. A method of optimizing performance of a redundant array of individual solid state drives comprising:

communicating, by a controller connected to the array, with said individual solid state drives; and

coordinating, by said controller connected to the array, garbage collection activities of each of the individual solid-state drives.

11. The method of claim 10, further including:

requesting said individual solid state drives, by the controller, to initiate the garbage collection activities, wherein garbage collection cycles begin on all the solid state drives in the array at a scheduled time.

12. The method of claim 10, wherein the communicating step includes one or more of said individual solid state drives notifying said controller of a need to perform garbage collection.

13. The method of claim 12, wherein the step of coordinating includes said controller requesting all of said individual solid state drives to begin garbage collection cycle.

14. The method of claim 12, wherein the step of coordinating includes said controller delaying until another notification from another of said individual solid state drives before requesting all of said individual solid state drives to begin garbage collection cycle.

15. The method of claim 10, further including:

collecting periodically, by the controller connected to the array, information from said individual solid state drives, and the step of coordinating further includes coordinating based on said collected information.

16. The method of claim 15, wherein the step of collecting includes said controller pulling one or more of said individual solid state drives for said information, said individual solid state drives pushing said information to said controller, or combinations thereof.

17. The method of claim 10, further including:

identifying one or more points in input/output stream that are expected to have idle periods, the input/output stream associated with the solid state drives; and

initiating a global garbage collection activity to be performed during the identified one or more points.

18. A non-transitory computer readable storage medium storing a program of instructions executable by a machine to perform a method of optimizing performance of a redundant array of individual solid state drives, the method comprising:

communicating, by a controller connected to the array, with said individual solid state drives; and

coordinating, by said controller connected to the array, garbage collection activities of each of the individual solid-state drives using a controller connected to the array.

19. The non-transitory computer readable storage medium of claim 18, further including:

requesting said individual solid state drives, by the controller, to initiate the garbage collection activities, wherein garbage collection cycles begin on all the solid state drives in the array at a scheduled time.

20. The non-transitory computer readable storage medium of claim 18, wherein the communicating step includes one or more of said individual solid state drives notifying said controller of a need to perform garbage collection.

21. The non-transitory computer readable storage medium of claim 20, wherein the step of coordinating includes said controller requesting all of said individual solid state drives to begin garbage collection cycle.

22. The non-transitory computer readable storage medium of claim 20, wherein the step of coordinating includes said controller delaying until another notification from another of said individual solid state drives before requesting all of said individual solid state drives to begin garbage collection cycle.

23. The non-transitory computer readable storage medium of claim 18, further including:

collecting periodically, by the controller connected to the array, information from said individual solid state drives, and the step of coordinating further includes coordinating based on said collected information.

24. The non-transitory computer readable storage medium of claim 23, wherein the step of collecting includes said controller pulling one or more of said individual solid state drives for said information, said individual solid state drives pushing said information to said controller, or combinations thereof.

25. The non-transitory computer readable storage medium of claim 18, further including:

identifying one or more points in input/output stream that are expected to have idle periods, the input/output stream associated with the solid state drives; and

initiating a global garbage collection activity to be performed during the identified one or more points.