Clustered raid assimilation management

• US 9,170,746 B2
• Filed: 01/07/2014
• Issued: 10/27/2015
• Est. Priority Date: 01/07/2014
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Sundaram, Rajesh, Baddepudi, Bharat
• Primary Examiner(s): NGUYEN, HIEP T
• Application Number:
  US14/149,488
• Publication Number:
  US 20150193168A1
• Time in Patent Office: 658 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/2092 Techniques of failing over between control units
• CPC Class Codes: G06F 2201/805 Real-time
• CPC Class Codes: G06F 2201/82 Solving problems relating to consistency ensuring consistency in mirrored systems G06F11/2064
• CPC Class Codes: G06F 3/0604 Improving or facilitating administration, e.g. storage management
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0632 by initialisation or re-initialisation of storage systems
• CPC Class Codes: G06F 3/0634 by changing the state or mode of one or more devices
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

In one embodiment, a node is a member of a cluster having a plurality of nodes, where each node is coupled to one or more storage arrays of solid state drives (SSDs) that serve as main storage. The node executed a storage input/output (I/O) stack having a redundant array of independent disks (RAID) layer that organizes the SSDs within the one or more storage arrays as one or more RAID groups. Configuration information is stored as a cluster database. The configuration information identifies (i) one or more RAID groups associated with an extent store, (ii) SSDs within each RAID group, and (iii) an identification of a node that owns the extent store. The cluster database is stored separate and apart from the main storage.

Claims

1. A system comprising:

a central processing unit (CPU) of a node of a cluster having a plurality of nodes, each node coupled to one or more storage arrays of solid state drives (SSDs) configured to serve as main storage for the nodes of the cluster;

a memory coupled to the CPU and configured to store a storage input/output (I/O) stack having a redundant array of independent disks (RAID) layer executable by the CPU, the RAID layer configured to organize the SSDs within the one or more storage arrays as one or more RAID groups; and

a service storage device coupled to the CPU and configured to store configuration information as a cluster database, the configuration information identifying (i) the one or more RAID groups associated with an extent store, (ii) the SSDs within each RAID group, and (iii) an identification of a node that owns the extent store, the cluster database maintained separate and apart from the main storage to enable the node to access the configuration information without dependency on the main storage and to enable any node of the cluster to access a list of the SSDs in the extent store to perform assimilation and determine ownership of the extent store.

2. The system of claim 1 wherein the cluster database is maintained on each node of the cluster as a local SSD.

3. The system of claim 1 wherein the cluster database replaces ownership of the SSDs with ownership of the extent store.

4. The system of claim 1 wherein the extent store is defined by a RAID configuration topology tree structure using a multi-level hierarchy.

5. The system of claim 4 wherein a first level of the RAID configuration topology tree structure identifies SSDs, a second level of the tree structure identifies a particular RAID group of the one or more RAID groups to which the identified SSDs belong and a third level of the tree structure identifies the extent store to which the particular RAID group belongs.

6. The system of claim 1 wherein the cluster database has properties of consistency, durability and ensuring a quorum of node members with respect to failover redundancy.

7. The system of claim 6 wherein the property of ensuring a quorum of node members comprises a property of ensuring a quorum of active members when updating and servicing a RAID group.

8. A method comprising:

executing, by a node of a cluster having a plurality of nodes where each node is coupled to one or more storage arrays of solid state drives (SSDs) that serve as main storage, a storage input/output (I/O) stack, the storage I/O) stack having a redundant array of independent disks (RAID) layer that organizes the SSDs within the one or more storage arrays as one or more RAID groups; and

storing configuration information as a cluster database, the configuration information identifying

(i) the one or more RAID groups associated with an extent store,

(ii) the SSDs within each RAID group, and

(iii) an identification of a node that owns the extent store,

wherein the cluster database is stored separate and apart from the main storage.

9. The method of claim 8, further comprising:

maintaining the cluster database on each node of the cluster as a local SSD.

10. The method of claim 8 further comprising:

replacing ownership of the SSDs with ownership of the extent store.

11. The method of claim 8 further comprising:

defining the extent store by a RAID configuration topology tree structure using a multi-level hierarchy.

12. The method of claim 11 wherein a first level of the RAID configuration topology tree structure identifies the SSDs, a second level of the tree structure identifies a particular RAID group of the one or more RAID groups to which the identified SSDs belong and a third level of the tree structure identifies the extent store to which the particular RAID group belongs.

13. The method of claim 8 further comprising:

ensuring, in the cluster database, a quorum of node members with respect to failover redundancy.

14. The method of claim 13 wherein the ensuring comprises:

ensuring, in the cluster database, a quorum of active members when updating and servicing a RAID group.

15. A non-transitory computer readable medium including program instructions for execution on one or more processors, the program instructions when executed operable to:

provide a storage input/output (I/O) stack, the storage I/O stack having a redundant array of independent disks (RAID) layer that organizes solid state drives (SSDs) of one or more storage arrays of solid state drives (SSDs) that serve as main storage for a cluster having a plurality of nodes, the organization to cause the SSDs to operate as one or more RAID groups; and

store configuration information as a cluster database, the configuration information identifying

(i) the one or more RAID groups associated with an extent store,

(ii) the SSDs within each RAID group, and

(iii) an identification of a node that owns the extent store,

wherein the cluster database is stored separate and apart from the main storage.

16. The non-transitory computer readable medium of claim 15 wherein the cluster database is maintained on each node of the cluster as a local SSD.

17. The non-transitory computer readable medium of claim 15 wherein the extent store is defined by a RAID configuration topology tree structure using a multi-level hierarchy.

18. The non-transitory computer readable medium of claim 17 wherein a first level of the RAID configuration topology tree structure identifies the SSDs, a second level of the tree structure identifies a particular RAID group to which the identified SSDs belong and a third level of the tree structure identifies the extent store to which the particular RAID group belongs.

19. The non-transitory computer readable medium of claim 15 wherein the cluster database has properties of consistency, durability and ensuring a quorum of node members with respect to failover redundancy.

20. The non-transitory computer readable medium of claim 19 wherein the property of ensuring a quorum of node members comprises a property of ensuring a quorum of active members when updating and servicing a RAID group.