Pool-level storage management

Patent metadata

Field	Value
Publication number	US-11853583-B2
Application number	US-202117521552-A
Country	US
Kind code	B2
Filing date	Nov 8, 2021
Priority date	Sep 13, 2018
Publication date	Dec 26, 2023
Grant date	Dec 26, 2023

Abstract

A method of managing data storage using a management device that includes determining respective status information for a plurality of storage devices, and calculating, based on the status information, a respective cost for each of the plurality of storage devices using a cost function that includes one or more parameters including at least one of: a program/erase (P/E) parameter, a block error state parameter, a block error level parameter, and a workload parameter. The method further includes selecting a destination storage device of the plurality of storage devices based on at least some of the calculated costs, and writing data to the destination storage device.

First claim

What is claimed is:

1. A method of managing data storage using a management device, comprising:
selecting a source storage device of a plurality of storage devices, wherein the plurality of storage devices comprise one or more solid state drives (SSDs);
reading data from the source storage device;
determining that the read data is corrupted;
selecting a destination storage device of the plurality of storage devices based on a cost for the destination storage device calculated using a cost function;
determining error-corrected data or error-free data based on the corrupted read data; and
writing the error-corrected data or error-free data to the selected destination storage device,
wherein the cost function includes two or more SSD parameters including at least a block error state parameter, and a block error level parameter.

2. The method of claim 1 , wherein determining error-free data based on the corrupted read data comprises referencing an erasure code or RAID stripe.

3. The method of claim 1 , further comprising:
calculating a respective cost for each of the plurality of storage devices using the cost function; and
determining that the cost for the destination storage device is the lowest or the highest of the costs for the plurality of storage devices.

4. The method of claim 1 , further comprising:
determining respective status information for the plurality of storage devices; and
calculating, based on the status information, a respective cost for each SSD of the plurality of storage devices using the cost function.

5. The method of claim 4 , wherein determining the respective status information is performed responsive to a periodic trigger.

6. The method of claim 1 , further comprising performing a data migration process, wherein the data to be migrated is stored on the source storage device of the plurality of storage devices.

7. The method of claim 6 , wherein the data migration process comprises:
selecting the source storage device for analysis;
determining that the source storage device is not reliable; and
determining to migrate the source storage device.

8. The method of claim 7 , wherein determining that the source storage device is not reliable is based on determining that the calculated cost of the source storage device is at, above, or below a cost threshold.

9. The method of claim 6 , wherein the data migration process is triggered according to a refresh protocol to reduce the likelihood of read disturb.

10. The method of claim 1 , further comprising receiving, from an external device, a request to store the data, wherein selecting the destination storage device is performed responsive to receiving the request to store the data.

11. The method of claim 1 , further comprising:
selecting two or more destination storage devices of the plurality of storage devices including the storage device, based on the at least some of the calculated costs; and
writing the data to the selected two or more destination storage devices.

12. The method of claim 11 , further comprising determining a ranking for the plurality of storage devices based on the calculated costs, and wherein selecting the destination storage device comprises selecting the highest ranked storage device of the plurality of storage devices.

13. The method of claim 1 , wherein the cost function includes one or more respective weights for the one or more parameters.

14. The method of claim 13 , further comprising receiving a user input and determining the one or more respective weights based on the user input.

15. The method of claim 1 , further comprising determining that source storage device is unreliable, wherein selecting the destination storage device is performed responsive to determining that the source storage device is unreliable.

16. The method of claim 15 , wherein determining that the source storage device is unreliable includes determining a cost for the source storage device which is calculated using the cost function.

17. The method of claim 15 , wherein determining that the source storage device is unreliable includes determining a degree and/or type of corruption of the read data.

18. The method of claim 15 , wherein determining that the source storage device is unreliable includes determining one or more of a frequency or likelihood of read data errors, programming errors or bad block errors for the source storage device.

Assignees

• Kioxia Corp

Inventors

• Klein Yaron

Classifications

• G06F3/0649 Primary

  Lifecycle management · CPC title

• G06F3/0605

  by facilitating the interaction with a user or administrator · CPC title

• G06F3/0614

  Improving the reliability of storage systems · CPC title

• G06F3/0683

  Plurality of storage devices · CPC title

• G06F11/1076

  Parity data used in redundant arrays of independent storages, e.g. in RAID systems · CPC title