Overlapping raid groups

• US 10,515,701 B1
• Filed: 10/31/2017
• Issued: 12/24/2019
• Est. Priority Date: 10/31/2017
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Aster, Radek, Bernat, Andrew R., Duan, Yunpeng, Scholbrock, Benjamin, Chang, Tsu-Hao
• Primary Examiner(s): Khan, Masud K
• Application Number:
  US15/799,979
• Time in Patent Office: 784 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 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 16/1752 based on file chunks
• CPC Class Codes: G06F 2212/7201 Logical to physical mapping or translation of blocks or pages
• CPC Class Codes: G06F 2212/7202 Allocation control and policies
• CPC Class Codes: G06F 2212/7205 Cleaning, compaction, garbage collection, erase control
• CPC Class Codes: G06F 2212/7207 management of metadata or control data
• CPC Class Codes: G06F 2212/7208 Multiple device management, e.g. distributing data over multiple flash devices
• 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 3/064 Management of blocks
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0652 Erasing, e.g. deleting, data cleaning, moving of data to a wastebasket
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: G11C 16/16 for erasing blocks, e.g. arrays, words, groups

Abstract

A method of using boot-time metadata in a storage system is provided. The method includes writing a fragmentation stride to a solid-state storage device of the storage system, the fragmentation stride defining a granularity on which fragmentation of erase blocks of the solid-state storage device occurs. The method includes allocating portions of erase blocks for at least one process in the storage system, in accordance with the fragmentation stride and writing boot up metadata at offsets that are based on the fragmentation stride, in the solid-state storage device.

Claims

1. A method of using boot-time metadata in a storage system, comprising:

determining a fragmentation stride based on a granularity on which fragmentation of erase blocks of a plurality of solid-state storage devices of the storage system occurs, wherein at least two of the plurality of solid-state storage devices have different erase block sizes;

allocating portions of erase blocks of one or more of the plurality of solid-state storage devices for at least one process in the storage system, in accordance with the fragmentation stride; and

writing boot up metadata at offsets that are based on the fragmentation stride, in the one or more of the plurality of solid-state storage devices.

2. The method of claim 1, further comprising:

accessing the boot up metadata at one or more of the offsets that are based on the fragmentation stride, at boot time of the storage system.

3. The method of claim 1, further comprising:

determining the fragmentation stride as a largest common factor of a plurality of erase block sizes of the plurality of solid-state storage devices of the storage system.

4. The method of claim 1, wherein the fragmentation stride equals 1 MB (megabyte).

5. The method of claim 1, wherein the allocating the portions of the erase blocks comprises:

allocating a first plurality of portions of erase blocks to at least a first process expecting to write short-lived data or metadata; and

allocating a second plurality of portions of erase blocks to at least a second process expecting to write long-lived data or metadata.

6. The method of claim 1, further comprising:

identifying a portion of live data or metadata in the plurality of solid-state storage devices;

requesting at least one of the portions of erase blocks for long-lived data or metadata; and

moving the portion of live data or metadata to the at least one of the portions of erase blocks for the long-lived data or metadata during garbage collection.

7. The method of claim 1, further comprising:

requesting at least one of the portions of erase blocks, with a hint as to whether data or metadata for the at least one of the portions of erase blocks is expected short-term, medium-term or long-term.

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:

writing a fragmentation stride to headers for device drivers of a plurality of solid-state storage devices of a storage system, to define granularity for fragmentation of erase blocks, wherein at least two of the plurality of solid-state storage devices have different erase block sizes;

allocating portions of erase blocks of one or more of the plurality of solid-state storage devices in accordance with the defined granularity, for writing data or metadata; and

writing metadata for boot up, to the one or more of the plurality of solid-state storage devices, at one or more offsets based on the fragmentation stride.

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

reading the metadata for boot up, at the one or more offsets based on the fragmentation stride, at boot time for the storage system.

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

determining a largest common factor of erase block sizes of the plurality of solid-state storage devices of the storage system, as the fragmentation stride.

11. The computer-readable media of claim 8, wherein the allocating the portions of the erase blocks comprises:

allocating portions of erase blocks from a first plurality of erase blocks, for short-term data or metadata; and

allocating portions of erase blocks from a second plurality of erase blocks, for long-term data or metadata.

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

identifying live data or metadata and obsoleted data or metadata in the plurality of solid-state storage devices, for garbage collection;

requesting allocation of one or more portions of erase blocks for long-term data or metadata as distinct from portions of erase blocks for short-term data or metadata;

transferring live data or metadata from an erase block to the one or more portions of erase blocks; and

erasing the erase block.

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

determining that data or metadata to be written is expected short-term, medium-term or long-term; and

requesting allocation of one or more portions of erase blocks that are specific to the expected short-term, medium-term or long-term.

14. A storage system, comprising:

a plurality of solid-state storage devices having memory with erase blocks, wherein at least two of the plurality of solid-state storage devices have different erase block sizes; and

at least one processor, operatively coupled to the plurality of solid-state storage devices, to;

determine a fragmentation stride for the plurality of solid-state storage devices defining a granularity of portion sizes of the erase blocks;

allocate portions of erase blocks in accordance with the fragmentation stride for writing data or metadata; and

write metadata for boot up, at one or more offsets based on the fragmentation stride, in each of the plurality of solid-state storage devices.

15. The storage system of claim 14, wherein the at least one processor is further to:

read the metadata for boot up, at boot time, in accordance with the one or more offsets based on the fragmentation stride.

16. The storage system of claim 14, wherein the at least one processor is further to:

determine a largest common factor of erase block sizes of the plurality of solid-state storage devices of the storage system, wherein the fragmentation stride comprises the largest common factor of the erase block sizes.

17. The storage system of claim 14, wherein to write the fragmentation stride comprises to write a value indicating 1 MB as the fragmentation stride.

18. The storage system of claim 14, wherein to allocate the portions of the erase blocks, the at least one processor to:

allocate a first plurality of portions of first erase blocks for short-term data or metadata; and

allocate a second plurality of portions of second erase blocks for long-term data or metadata.

19. The storage system of claim 14, wherein the at least one processor is further to:

determine that a portion of data or metadata in one of the plurality of solid-state storage devices is live;

request allocation of at least one of the portions of erase blocks for long-term data;

relocate the portion of data or metadata that is determined live from one or more erase blocks to the at least one of the portions of erase blocks for long-term data; and

erase the one or more erase blocks.

20. The storage system of claim 14, wherein the at least one processor is further to:

determine an expectation of short-term, medium-term or long-term, for data or metadata to be written to one of the portions of erase blocks, wherein the portions of erase blocks are allocated in accordance with the expectation.