Granted Patent B2

US 12,430,059 · App. 17/985,710 · Granted Sep 30, 2025

Inventors: Yang Sup Lee (San Jose, CA); Frank Tuzzolino (San Jose, CA); Douglas Lother (El Dorado Hills, CA); Casey Golliher (Mountain View, CA)

Assignee: PURE STORAGE, INC.

G06F3/0653G06F3/0614G06F3/0629G06F3/067G06F3/0683G06N20/00

Abstract

Claims (39)

1. A storage system comprising:

a storage controller operatively coupled to storage drives of the storage system, the storage controller configured to:

determine one or more characteristics associated with a first storage drive; and

provide tuning information to one or more other storage drives of the plurality of storage drives comprising the one or more characteristics, wherein providing the tuning information causes the one or more other storage drives to use the tuning information for data storage and retrieval and wherein the one or more other storage drives comprise one or more respective drive controllers separate from a first driver controller of the first storage drive.

2. The storage system of claim 1 , wherein the tuning information comprises voltage threshold tuning for one or more memory cells of the first storage drive.

3. The storage system of claim 1 , wherein the storage controller is a central storage controller that is external to the plurality of storage drives.

4. The storage system of claim 1 , wherein the storage controller is a local storage controller of the plurality of storage drives.

5. The storage system of claim 1 , wherein the storage controller is further configured to:

provide the tuning information and the one or more characteristics as inputs to a machine learning model, the machine learning model to determine optimized tuning information for storage drives based on corresponding characteristics of the plurality of storage drives.

6. The storage system of claim 5 , wherein the storage controller is further configured to:

identify a new storage drive to be added to the storage system;

receive, from the machine learning model, the optimized tuning information for the new storage drive; and

cause the new storage drive to apply the optimized tuning information.

7. The storage system of claim 5 , wherein the machine learning model is executed by a cloud computing system.

8. A method, comprising:

determining one or more characteristics associated with a first storage drive of a plurality of storage drives of a storage system; and

providing the tuning information to one or more other storage drives of the plurality of storage drives comprising the one or more characteristics, wherein providing the tuning information causes the one or more other storage drives to apply the tuning information for data storage and retrieval and wherein the one or more other storage drives comprise one or more respective drive controllers separate from a first driver controller of the first storage drive.

9. The method of claim 8 , wherein the tuning information comprises voltage threshold tuning for one or more memory cells of the first storage drive.

10. The method of claim 8 , wherein the storage controller is a central storage controller that is external to the plurality of storage drives.

11. The method of claim 8 , wherein the storage controller is a local storage controller of the first storage drive.

12. The method of claim 8 , further comprising:

providing the tuning information and the one or more characteristics as inputs to a machine learning model, the machine learning model to determine optimized tuning information for storage drives based on corresponding characteristics of the plurality of storage drives.

13. The method of claim 12 , further comprising:

identifying a new storage drive to be added to the storage system;

receiving, from the machine learning model, the optimized tuning information for the new storage drive; and

causing the new storage drive to apply the optimized tuning information.

14. The method of claim 12 , wherein the machine learning model is executed by a cloud computing system.

15. A non-transitory computer readable storage medium storing instructions, which when executed, cause a processing device of a storage controller to:

determine one or more characteristics associated with a first storage drive of a plurality of storage drives of a storage system; and

provide the tuning information to one or more other storage drives of the plurality of storage drives comprising the one or more characteristics, wherein providing the tuning information causes the one or more other storage drives to apply the tuning information for data storage and retrieval and wherein the one or more other storage drives comprise one or more drive respective controllers separate from a first driver controller of the first storage drive.

16. The non-transitory computer readable storage medium of claim 15 , wherein the tuning information comprises voltage threshold tuning for one or more memory cells of the first storage drive.

17. The non-transitory computer readable storage medium of claim 15 , wherein the storage controller is a central storage controller that is external to the plurality of storage drives.

18. The non-transitory computer readable storage medium of claim 15 , wherein the storage controller is a local storage controller of the first storage drive.

19. The non-transitory computer readable storage medium of claim 15 , wherein the processing device is further to:

provide the tuning information and the one or more characteristics as inputs to a machine learning model, the machine learning model to determine optimized tuning information for storage drives based on corresponding characteristics of the plurality of storage drives.

20. The non-transitory computer readable storage medium of claim 19 , wherein the processing device is further to:

identify a new storage drive to be added to a storage system;

receive, from the machine learning model, the optimized tuning information for the new storage drive; and

cause the new storage drive to apply the optimized tuning information.

Continuity (2)

References Cited (400)