Geometry based, space aware shelf/writegroup evacuation

• US 10,203,903 B2
• Filed: 10/31/2016
• Issued: 02/12/2019
• Est. Priority Date: 07/26/2016
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Feigin, Boris, Kleinerman, Andrew, Tumanova, Svitlana, Vohra, Taher, Wang, Xiaohui
• Primary Examiner(s): Chan, Tracy C.
• Application Number:
  US15/339,757
• Publication Number:
  US 20180032280A1
• Time in Patent Office: 834 Days
• Field of Search: None
• 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 2212/7205 Cleaning, compaction, garbage collection, erase control
• CPC Class Codes: G06F 3/0607 by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays

Abstract

A method for migration of data is provided. The method includes triggering a rebuild of data according to a first migration mechanism from a first storage drive to a second storage drive. Monitoring space utilization associated with the second storage drive, and adaptively switching the migration of the data from the first migration mechanism to a second migration mechanism based on the monitoring.

Claims

1. A method of data migration within a storage system, performed by the storage system, comprising:

determining an amount of storage memory space that is used or available in a plurality of portions of storage memory of a storage system, relative to a threshold value, the determining comprising;

comparing a value associated with an imbalance of one or more of the portions of storage memory with the threshold value;

adjusting relative utilization of the first migration mechanism and the second migration mechanism based on the comparing;

selecting relative utilization of a first migration mechanism and a second migration mechanism, to perform within the storage memory, based on the determining; and

performing data migration within the storage memory, in accordance with the selecting.

2. The method of claim 1, further comprising:

repeating the determining and the selecting during the performing; and

adjusting the relative utilization of the first migration mechanism and the second migration mechanism.

3. The method of claim 1, wherein each of the plurality of portions of storage memory comprises a write group.

4. The method of claim 1, wherein each of the plurality of portions of storage memory comprises a solid state drive (SSD).

5. The method of claim 1, wherein the plurality of portions of storage memory comprises an entire amount of storage memory of the storage system.

6. The method of claim 1, wherein the performing the data migration comprises:

evacuating a portion of the storage memory.

7. The method of claim 1, wherein the first migration mechanism comprises a data rebuild and the second migration mechanism comprises a garbage collection operation.

8. The method of claim 1, wherein the determining comprises:

generating the value associated with the imbalance.

9. A tangible, non-transitory, computer-readable media having instructions thereupon which, when executed by a processor, cause the processor to perform a method comprising:

determining an amount of storage memory space that is used or available in a plurality of portions of storage memory of a storage system, relative to a threshold value;

selecting relative utilization of a first migration mechanism and a second migration mechanism, to perform within the storage memory, based on the determining;

performing data migration within the storage memory, in accordance with the selecting; and

switching, in a dynamic manner in accordance with the selecting, between a RAID (redundant array of independent disks) rebuild as the first migration mechanism when the determining indicates balance, and a garbage collection data move as the second migration mechanism when the determining indicates imbalance.

10. The tangible, non-transitory, computer-readable media of claim 9, wherein the method further comprises:

repeating the determining and the selecting during the performing; and

adjusting the relative utilization of the first migration mechanism and the second migration mechanism.

11. The tangible, non-transitory, computer-readable media of claim 9, wherein:

the selecting includes selecting a greater amount or rate of migrating live and dead data when the determining indicates balance across write groups, and selecting a greater amount or rate of migrating solely live data when the determining indicates imbalance across the write groups.

12. The tangible, non-transitory, computer-readable media of claim 9, wherein:

the selecting includes selecting a greater amount or rate of migrating live and dead data when the determining indicates balance across solid-state drives, and selecting a greater amount or rate of migrating solely live data when the determining indicates imbalance across the solid-state drives.

13. The tangible, non-transitory, computer-readable media of claim 9, wherein the determining comprises:

generating a value associated with an imbalance of one or more of the portions of storage memory;

comparing the value with the threshold value; and

adjusting relative utilization of the first migration mechanism and the second migration mechanism based on the comparing.

14. A storage system, comprising:

storage memory;

one or more processors;

a space monitor, implemented in hardware, firmware or software executing on the one or more processors, the space monitor configurable to track storage memory space used for writing or available for writing in a plurality of portions of the storage memory; and

a migration module, implemented in hardware, firmware or software executing on the one or more processors, the migration module configurable to vary relative amounts or rates of a first migration mechanism and a second migration mechanism within the storage system, with varying the relative amounts or rates responsive to the space monitor tracking an amount of imbalance of the storage memory space used or available in the plurality of the portions of storage memory, the migration module configurable to perform a rebuild, an evacuation, or a targeted remap, using data migration switching between migrating live data and dead data and migrating only live data in a dynamic manner based on the amount of imbalance during data migration.

15. The storage system of claim 14, further comprising:

the one or more processors including at least two processors arranged as a high-availability controller; and

each of the plurality of portions of the storage memory including a write group that has a specified level of redundancy or error correction for data and acts as a failure domain, wherein the amount of imbalance is an imbalance across write groups.

16. The storage system of claim 14, further comprising:

a plurality of blades each having a node and having solid-state memory as one of the plurality of portions of the storage memory, with multiple nodes coupled together as a storage cluster, wherein the amount of imbalance is an imbalance across blades.

17. The storage system of claim 14, wherein:

each of the plurality of portions of storage memory comprises a solid-state drive or solid-state storage; and

the amount of imbalance is an imbalance across solid-state drives or solid-state storages.

18. The storage system of claim 14, wherein the first migration mechanism is a RAID (redundant array of independent disks) rebuild and the second migration is a garbage collection data move.