Storage system with elective garbage collection to reduce flash contention

• US 10,366,004 B2
• Filed: 11/22/2016
• Issued: 07/30/2019
• Est. Priority Date: 07/26/2016
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Feigin, Boris, Lee, Robert, Tumanova, Svitlana, Vohra, Taher
• Primary Examiner(s): Dudek, Jr., Edward J
• Application Number:
  US15/359,225
• Publication Number:
  US 20180074951A1
• Time in Patent Office: 980 Days
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 11/3034 where the computing system component is a storage system, e.g. DASD based or network based digital input from or digital output to record carriers G06F3/06; digital recording or reproducing G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 11/3485 for I/O devices
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 2212/7205 Cleaning, compaction, garbage collection, erase control
• CPC Class Codes: G06F 3/0605 by facilitating the interaction with a user or administrator
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/0652 Erasing, e.g. deleting, data cleaning, moving of data to a wastebasket
• 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 elective garbage collection in storage memory, performed by a storage system is provided. The method includes monitoring storage space available in each of a plurality of portions of storage memory of a storage system and detecting an imbalance in the storage space available across the plurality of portions of storage memory. The method includes performing garbage collection to rebalance the space available across the plurality of portions of storage memory, responsive to the detecting. A storage system is also provided.

Claims

1. A method for elective garbage collection in storage memory, performed by a storage system, comprising:

monitoring storage space available in each of a plurality of portions of storage memory of a storage system;

detecting an imbalance in the storage space available across the plurality of portions of storage memory; and

selecting between a RAID rebuild and a garbage collection move to perform for data migration, based on the detecting, wherein both the RAID rebuild and the garbage collection move are configurable to stripe data across each of the plurality of portions of storage memory.

2. The method of claim 1, further comprising:

detecting addition of a further portion of storage memory to the storage system, wherein the imbalance results from the addition of the further portion of storage memory, and wherein such rebalancing achieves linearity of performance by preventing bottlenecks as the storage system is scaled by the addition of the further portion of storage memory.

3. The method of claim 1, further comprising:

identifying addition of a further portion of storage memory to the storage system; and

declaring an entirety of the plurality of portions of storage memory as candidate memory for the garbage collection, responsive to the identifying the addition of the further portion of storage memory, so that the garbage collection reads from the plurality of portions of storage memory and writes across the plurality of portions of storage memory and the further portion of storage memory.

4. The method of claim 1, wherein:

each of the plurality of portions of storage memory comprises a storage pack, one or more solid-state drives, or a blade having solid-state memory; and

detecting the imbalance is a result of identifying a difference in distribution of storage space available among the plurality of portions of storage memory.

5. The method of claim 1, wherein the performing garbage collection comprises:

reading live data from garbage collection candidate portions of the storage memory; and

striping the live data across the plurality of portions of storage memory including at least one portion of storage memory that has more storage space available than the garbage collection candidate portions of the storage memory.

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

continuing the monitoring and adjusting the selecting based on the monitoring;

and wherein striping data across each of the plurality of portions of storage memory includes a portion of the storage memory associated with the selecting.

7. The method of claim 1, further comprising:

initiating garbage collection responsive to detecting overall storage space available in the storage system being below a threshold.

8. 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, for each of a plurality of portions of storage memory of a storage system, storage space available;

detecting that storage space available across the plurality of portions of storage memory has an imbalance; and

selecting between a RAID rebuild and a garbage collection move to perform for data migration, based on the detecting, wherein both the RAID rebuild and the garbage collection move are configurable to stripe data across each of the plurality of portions of storage memory.

9. The computer-readable media of claim 8, wherein the method further comprises:

adding a further portion of storage memory to the storage system, with the detecting and the selecting responsive to the adding, and wherein striping data across each of the plurality of portions of storage memory includes a portion of the storage memory associated with the selecting.

10. The computer-readable media of claim 8, wherein the method further comprises:

identifying a further portion of storage memory is added to the storage system, wherein the detecting the imbalance is based on the identifying the further portion of storage memory is added and wherein an entirety of the plurality of portions of storage memory is candidate memory for the garbage collection.

11. The computer-readable media of claim 8, wherein the determining the storage space available for each of the plurality of portions of storage memory and the detecting the imbalance comprises identifying a difference in distribution of storage space available among the plurality of portions of storage memory.

12. The computer-readable media of claim 8, wherein the method further comprises:

striping live data, from garbage collection candidate portions of the storage memory, across the plurality of portions of storage memory in performance of the garbage collection, wherein at least one of the plurality of portions of storage memory has more storage space available than the garbage collection candidate portions of the storage memory.

13. The computer-readable media of claim 8, wherein the determining the storage space available for each of the plurality of portions of storage memory comprises:

polling a storage space detector so as to determine storage space used in each of the plurality of portions of storage memory.

14. A storage system, comprising:

one or more processors;

a plurality of portions of storage memory;

a storage space detector, configurable to track storage space available in each of the plurality of portions of storage memory and configurable to select between one of RAID rebuild and a garbage collection move to perform for data migration, based on the tracked storage space available; and

a garbage collection module, configurable to perform garbage collection so as to rebalance the storage space available across the plurality of portions of storage memory, responsive to identifying an imbalance across the plurality of portions of storage memory based on results from the storage space detector, wherein both the RAID rebuild and the garbage collection move are configurable to stripe data across each of the plurality of portions of storage memory.

15. The storage system of claim 14, wherein the storage space detector and the garbage collection module are further configurable to determine addition of a further portion of storage memory to the storage system, with the identifying the imbalance triggered by the addition of the further portion of storage memory.

16. The storage system of claim 14, wherein:

the one or more processors are configurable to determine addition of a further portion of storage memory to the storage system; and

the garbage collection module is further configurable to collect from the plurality of portions of storage memory as candidate memory for the garbage collection, and write across the plurality of portions of storage memory and the further portion of storage memory, restoring balance in the storage memory space available in the storage system.

17. The storage system of claim 14, wherein:

the plurality of portions of storage memory comprises a plurality of storage packs, solid-state drives, or blades having solid-state memory; and

identifying the imbalance is responsive to identifying a difference in distribution of storage space available among the plurality of portions of storage memory.

18. The storage system of claim 14, wherein the garbage collection module is configurable to read live data from a garbage collection candidate portion of the storage memory and stripe the live data across the plurality of portions of storage memory including a portion of storage memory that has more storage space available than the garbage collection candidate portion of the storage memory.

19. The storage system of claim 14, wherein the storage space detector continues to adjust between the RAID rebuild and the garbage collection based on the tracked storage space, and wherein striping data across each of the plurality of portions of storage memory includes a portion of the storage memory associated with selecting between the RAID rebuild and the garbage collection move.

20. The storage system of claim 14, wherein the one or more processors are configurable to poll the space detector module to determine overall storage space available in the storage system and to trigger the garbage collection responsive to determining the overall storage space available in the storage system is below a threshold.