Building Stable Storage Area Networks For Compute Clusters

• US 20190034094A1
• Filed: 09/13/2017
• Published: 01/31/2019
• Est. Priority Date: 07/31/2017
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Jibbe, Mahmoud K., Lang, Dean, Parnell, Joey, Rodine, Ryan, Briner, Joshua
• Granted Patent: US 10,521,127 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 12/0808 with cache invalidating means G06F12/0815 takes precedence
• CPC Class Codes: G06F 12/0868 Data transfer between cache memory and other subsystems, e.g. storage devices or host systems
• CPC Class Codes: G06F 12/0888 using selective caching, e.g. bypass
• CPC Class Codes: G06F 2212/1024 Latency reduction
• CPC Class Codes: G06F 2212/313 In storage device
• CPC Class Codes: G06F 2212/502 using adaptive policy
• CPC Class Codes: G06F 2212/621 Coherency control relating to peripheral accessing, e.g. from DMA or I/O device
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• 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 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

Systems and methods that result in a stable storage system are provided. In the storage system, the latency spikes may be reduced when multiple volumes are aggregated into transfer sets according to system characteristics. The storage system transfers ownership of volumes in each transfer set as a single transaction. In the storage system, connectivity between the host and the storage controller is re-established based on the connectivity in a physical transport layer and a single path. In the storage system, pre-mature failback is also avoided when ownership of volumes is transferred back to a preferred storage controller when the same number of paths existed between the host and the preferred storage controller before and after a failover operation. Further, the storage system generates connectivity reports that display connectivity paths between hosts, storage controllers, and volumes.

Claims

1. A method comprising:

aggregating volumes in a plurality of volumes into a transfer set according to system characteristics;

discontinuing I/O requests to volumes in the transfer set;

transferring ownership of volumes in the transfer set to one or more storage controllers that have not owned the volumes in the transfer set; and

reinstating the I/O requests to the volumes in the transfer set.

2. The method of claim 1, wherein system characteristics include ownership of at least one storage controller that owns a volume of the plurality of volumes.

3. The method of claim 1, wherein system characteristics include media type of at least one storage device that is associated with a volume of the plurality of volumes.

4. The method of claim 1, wherein the system characteristics include Redundant Array of Independent/Inexpensive Disks (“

RAID”

) volume group membership.

5. The method of claim 1, wherein the system characteristics include workload type of applications that store data in the plurality of volumes.

6. The method of claim 1, further comprising:

changing a caching model for a cache that stores data associated with the volumes in the transfer set to a write through model.

7. The method of claim 1, further comprising:

flushing a cache that stores data associated with the volumes in the transfer set to underlying one or more storage devices.

8. The method of claim 1, wherein the aggregating occurs during a failback operation.

9. The method of claim 1, further comprising:

generating a connectivity report displaying the one or more storage controllers having ownership of the volumes in the transfer set.

10. A computing device comprising:

a memory containing machine readable medium comprising machine executable code having stored thereon instructions for performing a method of transferring ownership of volumes; and

a processor coupled to the memory, the processor configured to execute the machine executable code to cause the processor to;

aggregate a subset of volumes in a plurality of volumes into a transfer set;

change a caching model for at least one cache that stores data associated with the subset of volumes to a write-thru model;

flush the data in the at least one cache to underlying one or more storage devices;

discontinue I/O requests to the subset of volumes;

transfer ownership of the subset of the volumes in the transfer set; and

reinstate the I/O requests to the subset of volumes.

11. The computing device of claim 10, wherein system characteristics include ownership of at least one storage controller that owns a volume in the plurality of volumes.

12. The computing device of claim 10, wherein system characteristics include media type of at least one storage device that is associated with a volume in the plurality of volumes.

13. The computing device of claim 10, wherein the system characteristics include Redundant Array of Independent/Inexpensive Disks (“

RAID”

) volume group membership.

14. The computing device of claim 10, wherein the system characteristics include workload type of applications that store data in the plurality of volumes.

15. The computing device of claim 10, wherein the processor aggregates the plurality of volumes during a failback operation.

16. The computing device of claim 10, wherein the processor is further configured to execute the machine executable code to cause the processor to:

generate a connectivity report displaying the one or more storage controllers having ownership of volumes in the transfer set.

17. A non-transitory machine readable medium having stored thereon instructions for performing a method comprising machine executable code which when executed by at least one machine, causes the machine to:

determine a number of paths between a host and a preferred storage controller, wherein the preferred storage controller has ownership of a volume;

perform a failover operation, wherein the failover operation transfers ownership of the volume from the preferred storage controller to a non-preferred storage controller;

determine a number of paths between the host and the preferred storage controller subsequent to the failover operation;

compare the number of paths before the failover operation to the number of paths subsequent to the failover operation; and

transfer the ownership of the volume from the non-preferred storage controller to the preferred storage controller based on comparing the number of paths.

18. The computing device of claim 17, wherein the processor is configured to execute the machine executable code to:

identify a connectivity fault between the host and the preferred storage controller; and

wherein the failover operation occurs because of the connectivity fault.

19. The computing device of claim 17, wherein the transfer of ownership of the volume occurs when the number of paths between the host and the preferred storage controller before the failover operation is the same as the number of paths between the host and the preferred storage controller subsequent to the failover operation.

20. The computing device of claim 17, wherein the processor is configured to execute the machine executable code to cause the processor to:

identify an additional path between the host and the preferred storage controller prior to the transfer of the ownership of the volume from the non-preferred storage controller to the preferred storage controller