Hyperconverged storage system with distributable processing power

• US 10,379,763 B2
• Filed: 12/04/2017
• Issued: 08/13/2019
• Est. Priority Date: 06/04/2014
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Colgrove, John, Davis, John D., Hayes, John Martin, Lee, Robert
• Primary Examiner(s): Song, Hua J
• Application Number:
  US15/830,987
• Publication Number:
  US 20180101321A1
• Time in Patent Office: 617 Days
• Field of Search: 711103
• CPC Class Codes: G06F 11/1068 in sector programmable memories, e.g. flash disk G06F11/1072 takes precedence
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/108 Parity data distribution in semiconductor storages, e.g. in SSD
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 2201/845 Systems in which the redundancy can be transformed in increased performance
• CPC Class Codes: G06F 2212/7206 Reconfiguration of flash memory system
• CPC Class Codes: G06F 2212/7207 management of metadata or control data
• CPC Class Codes: G06F 3/06 Digital input from, or digital output to, record carriers , e.g. RAID, emulated record carriers or networked record carriers
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/0622 in relation to access
• CPC Class Codes: G06F 3/0637 Permissions
• CPC Class Codes: G06F 3/0655 Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays
• CPC Class Codes: H03M 13/1102 Codes on graphs and decoding on graphs, e.g. low-density parity check [LDPC] codes
• CPC Class Codes: H03M 13/1515 Reed-Solomon codes
• CPC Class Codes: H03M 13/154 Error and erasure correction, e.g. by using the error and erasure locator or Forney polynomial
• CPC Class Codes: H03M 13/3761 using code combining, i.e. using combining of codeword portions which may have been transmitted separately, e.g. Digital Fountain codes, Raptor codes or Luby Transform [LT] codes

Abstract

A method for managing processing power in a storage system is provided. The method includes providing a plurality of blades, each of a first subset having a storage node and storage memory, and each of a second, differing subset having a compute-only node. The method includes distributing authorities across the plurality of blades, to a plurality of nodes including at least one compute-only node, wherein each authority has ownership of a range of user data.

Claims

1. A system configurable for distributing processing power across a plurality of blades of a storage system, the system, comprising:

a plurality of blades, a first subset of blades of the plurality of blades including a storage node and storage memory, and a second subset of blades of the plurality of blades comprising a compute-only node;

a plurality of authorities in the system with each authority owning a range of data stored in the system, the plurality of authorities configurable to write erasure coded data representing the range of data across the first subset of blades of the plurality of blades.

2. The system of claim 1, wherein the plurality of blades are contained within multiple chassis.

3. The system of claim 1, wherein the plurality of authorities are distributed at least across the first subset of blades of the plurality of blades.

4. The system of claim 1, wherein an I/O request received for the external I/O processing is routed from one of the plurality of blades to another one of the plurality of blades, in accordance with the plurality of authorities.

5. The system of claim 1, wherein at least one of the plurality of authorities is relocatable from one of the plurality of blades to another of the plurality of blades.

6. The system of claim 1, wherein the plurality of authorities are distributed across the plurality of blades in proportion to an amount of RAM (random-access memory) on each of the plurality of blades.

7. The system of claim 1, wherein processing power for external I/O processing is distributed in accordance with one or more policies, agreements, service classes or multi-tenant service.

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

distributing a plurality of authorities in the storage system, across a plurality of blades of the storage system, wherein the plurality of blades comprises a first subset of blades of the plurality of blades including a storage node and storage memory, and a second subset of blades of the plurality of blades comprising a compute-only node, and wherein each of the plurality of authorities owns a range of data; and

writing erasure coded data representing the range of data across the first subset of blades of the plurality of blades, as directed by the plurality of authorities.

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

distributing the plurality of authorities at least across the first subset of blades of the plurality of blades.

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

routing an I/O request, received for external I/O processing, from one of the plurality of blades to another of the plurality of blades, in accordance with the plurality of authorities.

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

moving one or more of the plurality of authorities from one or more of the first subset of blades of the plurality of blades to the one or more of the second subset of blades of the plurality of blades.

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

distributing the plurality of authorities across at least a subset of the plurality of blades in proportion to RAM (random-access memory) on each blade of the subset of the plurality of blades, wherein the RAM comprises RAM, DRAM (dynamic random access memory) or NVRAM (nonvolatile random-access memory).

13. The computer-readable media of claim 8, wherein the method comprises determining I/O request processing and throughput in accordance with one or more policies, agreements, service classes or multi-tenant service.

14. A storage cluster, comprising:

a plurality of blades, each having a storage node and storage memory;

at least one compute blade, having a compute-only node and no storage memory; and

a plurality of authorities distributed in the storage cluster with each owning a range of data, the plurality of authorities configurable to write erasure coded data representing the range of data across the plurality of blades.

15. The storage cluster of claim 14, wherein the plurality of blades are contained within multiple chassis.

16. The storage cluster of claim 14, wherein processing power is distributed according to distribution of the plurality of authorities across the plurality of blades.

17. The storage cluster of claim 16, wherein the distributed processing power is configurable for routing an I/O request, received for external I/O processing, from one of the plurality of blades to a differing one of the plurality of blades, in accordance with the plurality of authorities.

18. The storage cluster of claim 16, wherein the distributed processing power is configurable for moving one or more of the plurality of authorities from one of the plurality of blades to a differing one of the plurality of blades.

19. The storage cluster of claim 16, wherein the distributed processing power is configurable for distributing the plurality of authorities across the plurality of blades and the at least one compute blade in proportion to an amount of RAM on each of the plurality of blades and the at least one compute blade.

20. The storage cluster of claim 16, wherein the distributed processing power for external I/O processing is allocatable in accordance with one or more policies, agreements, service classes or multi-tenant service.