Storage system and method for optimizing host-activated defragmentation and proactive garbage collection processes

Patent metadata

Field	Value
Publication number	US-11972149-B2
Application number	US-202217874555-A
Country	US
Kind code	B2
Filing date	Jul 27, 2022
Priority date	Jul 27, 2022
Publication date	Apr 30, 2024
Grant date	Apr 30, 2024

Abstract

A storage system is provided that performs a defragmentation operation or proactive garbage collection in its memory based on a command from a host. The command specifies which blocks in the memory should take part in the defragmentation operation by specifying a maximum amount of valid data that a block can have to qualify for defragmentation. That way, the storage system only performs defragmentation on those blocks that meet the validity criteria provided by the host. This can help improve performance of the storage system while reducing the degree of negative tradeoffs that may come with defragmentation or proactive garbage collection.

First claim

What is claimed is:

1. A storage system comprising:
a memory; and
a controller coupled with the memory and configured to:
receive, from a host, a command to perform host-activated defragmentation on the memory, wherein the command specifies a maximum block valid count threshold that is selected by the host after assessing a tradeoff between increased performance and reduction of memory lifetime associated with the maximum block valid count threshold;
perform defragmentation only to those blocks in the memory whose amount of valid data is not greater than the maximum block valid count threshold; and
receive, from the host, at least one subsequent command to perform host-activated defragmentation on the memory, wherein each subsequent command specifies a different maximum block valid count threshold;
wherein the host dynamically decreases the maximum block valid count threshold in each subsequent command as the host progressively prioritizes memory lifetime over performance.

2. The storage system of claim 1 , wherein the controller is further configured to inform the host that the storage system supports the command.

3. The storage system of claim 1 , wherein the controller is further configured to provide the host with an indication of a maximum block valid count threshold supported by the storage system.

4. The storage system of claim 1 , wherein the controller is further configured to provide the host with a recommendation of the maximum block valid count threshold.

5. The storage system of claim 1 , wherein the controller is further configured to provide the host with an indication of a number of blocks of the memory that are expected to be relocated in each of a plurality of maximum block valid count thresholds.

6. The storage system of claim 1 , wherein the defragmentation is performed during an idle time of the memory.

7. The storage system of claim 1 , wherein the command further indicates a storage pool in the memory to which the defragmentation operation is to be performed.

8. The storage system of claim 1 , wherein the maximum block valid count threshold is based on a counter value, a memory fullness, and/or performance statistics.

9. The storage system of claim 1 , wherein the memory comprises a three-dimensional memory.

10. In a storage system comprising a memory, a method comprising:
receiving, from a host, a command to perform host-activated defragmentation on the memory, wherein the command specifies a valid count criteria that is determined by host application usage after the host assesses a tradeoff between increased performance and reduction of memory lifetime associated with the valid count criteria;
performing defragmentation only on those blocks in the memory that satisfy the valid count criteria; and
receiving, from the host, at least one subsequent command to perform host-activated defragmentation on the memory, wherein each subsequent command specifies a different valid count criteria, and wherein the host dynamically decreases the valid count criteria in each subsequent command as the host progressively prioritizes memory lifetime over performance.

11. The method of claim 10 , further comprising informing the host that the storage system supports the command.

12. The method of claim 10 , further comprising providing the host with an indication of a maximum valid count criteria supported by the storage system.

13. The method of claim 10 , further comprising providing the host with a recommendation of the valid count criteria.

14. The method of claim 10 , further comprising providing the host with an indication of a number of blocks of the memory that are expected to be relocated in each of a plurality of valid count criteria.

15. The method of claim 10 , wherein the defragmentation is performed during an idle time of the memory.

16. The method of claim 10 , wherein the at least one subsequent command comprises a plurality of commands.

17. The method of claim 10 , wherein the command further indicates a storage pool in the memory to which the defragmentation operation is to be performed.

18. The method of claim 10 , wherein the valid count criteria is based on memory lifetime, memory health, a counter value, a memory fullness, and/or performance statistics.

19. A storage system comprising:
a memory;
means for receiving, from a host, a command to perform host-activated defragmentation on the memory, wherein the command specifies a selection of one of a predefined number of maximum block valid count threshold options that is selected by the host after assessing a tradeoff between increased performance and reduction of memory lifetime associated with each of the predefined number of maximum block valid count threshold options;
means for performing defragmentation only to those blocks in the memory whose amount of valid data is not greater than the selected one of the predefined number of maximum block valid count threshold options; and
means for receiving, from the host, at least one subsequent command to perform host-activated defragmentation on the memory, wherein each subsequent command specifies a different one of the predefined number of maximum block valid count threshold options, wherein the host dynamically selects a decreasing maximum block valid count threshold option in each subsequent command as the host progressively prioritizes memory lifetime over performance.

20. The storage system of claim 1 , wherein the maximum block valid count threshold is based on memory health.

Assignees

• Western Digital Tech Inc

Inventors

• Zilberstein Einav
• Sober Nadav
• Katz Omer

Classifications

• G06F3/0659 Primary

  Command handling arrangements, e.g. command buffers, queues, command scheduling · CPC title

• G06F3/0604

  Improving or facilitating administration, e.g. storage management · CPC title

• G06F3/0679

  Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP] · CPC title

• G06F12/0253

  Garbage collection, i.e. reclamation of unreferenced memory · CPC title

• G06F16/1724

  Details of de-fragmentation performed by the file system (saving storage space on storage systems G06F3/0608; management of blocks in storage devices G06F3/064) · CPC title