Method of balancing workloads in object storage system

• US 8,769,174 B2
• Filed: 08/29/2011
• Issued: 07/01/2014
• Est. Priority Date: 08/29/2011
• Status: Active Grant

• Current Assignee: VMware, Inc. (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Sokolinski, Ilia, Shih, Chiao-Chuan, Vaghani, Satyam B., Desai, Komal, Aswathanarayana, Tejasvi
• Primary Examiner(s): Alrobaye, Idriss N
• Assistant Examiner(s): Huson, Zachary K
• Application Number:
  US13/219,899
• Publication Number:
  US 20130055248A1
• Time in Patent Office: 1,037 Days
• Field of Search: None
• US Class Current: 710/74
• CPC Class Codes: G06F 2206/1012 Load balancing
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 3/0637 Permissions
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: H04L 67/1097 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]

Abstract

The storage system exports logical storage volumes that are provisioned as storage objects. These storage objects are accessed on demand by connected computer systems using standard protocols, such as SCSI and NFS, through logical endpoints for the protocol traffic that are configured in the storage system. When traffic through a particular protocol endpoint is busy, the logical storage volume undergoes a rebind process pursuant to which the logical storage volume is bound to a new protocol endpoint, and the binding to the old protocol endpoint is removed upon completion of all pending input-output commands issued through the old protocol endpoint.

Claims

1. In a storage system configured with protocol endpoints through which input-output commands (IOs) are received, a method for changing a binding of a logical storage volume from a first protocol endpoint to a second protocol endpoint, said method comprising:

while the logical storage volume is bound to the first protocol endpoint, binding the logical storage volume to the second protocol endpoint;

receiving IOs for the logical storage volume through both first and second protocol endpoints; and

unbinding the logical storage volume from the first protocol endpoint.

2. The method of claim 1, further comprising:

maintaining a data structure having an entry that stores information about the binding of the logical storage volume to the first protocol endpoint and the binding of the logical storage volume to the second protocol endpoint.

3. The method of claim 2, wherein the binding of the logical storage volume to the second protocol endpoint includes generating an identifier and adding an entry to the data structure that associates the second protocol endpoint with the identifier and the logical storage volume.

4. The method of claim 3, wherein the IOs issued to the logical storage volume that are received through the second protocol endpoint includes the identifier.

5. The method of claim 4, wherein the IOs issued to the logical storage volume that are received through the second protocol endpoint are SCSI compliant IOs and the identifier is a randomly generated identifier.

6. The method of claim 4, wherein the IOs issued to the logical storage volume that are received through the second protocol endpoint are NFS compliant IOs and the identifier is a file path to a file object in the NFS corresponding to the logical storage volume.

7. The method of claim 2, wherein each entry in the data structure includes a reference count that indicates the number of different processes of the computer system that are accessing the logical storage volume through the protocol endpoint corresponding to the entry.

8. The method of claim 7, wherein the reference count is decremented each time an unbind request is received from the computer system.

9. The method of claim 8, further comprising:

removing an entry in the data structure when the reference count associated with the entry decrements to zero.

10. The method of claim 1, wherein the storage system includes first and second storage subsystems that are physically separate and the first protocol endpoint is configured in the first storage subsystem and the second protocol endpoint is configured in the second storage subsystem.

11. In a computer system connected to a storage system configured with protocol endpoints through which input-output commands (IOs) to logical storage volumes provisioned in the storage system are received, a method for issuing IOs to a logical storage volume, the binding for which is to be changed from a first protocol endpoint to a second protocol endpoint, said method comprising:

sending a request for the logical storage volume to be bound to a new protocol endpoint and, in response to the request, receiving an identifier for the new protocol endpoint and an additional identifier;

issuing subsequent IOs to the logical storage volume through the new protocol endpoint; and

sending a request to unbind the logical storage volume from the first protocol endpoint.

12. The method of claim 11, wherein prior to sending the request to unbind, waiting for all pending IOs issued through the first protocol endpoint to complete.

13. The method of claim 11, further comprising:

maintaining a data structure having entries each corresponding to a block device instance of a logical storage volume to which IOs are to be issued, and indicating first and second identifiers that are to be included in the IOs.

14. The method of claim 13, wherein the data structure includes two entries with the same block device instance, one of the entries indicating the block device instance as active and the other of which indicating the block device instance as inactive.

15. The method of claim 14, wherein each entry in the data structure further indicates the number of IOs that have been issued to the corresponding block device instance but have not completed.

16. The method of claim 11, further comprising;

for each block device instance that is not active, sending a request to unbind the logical storage volume associated therewith when all IOs that have been issued to the block device instance have completed.

17. The method of claim 11, wherein the subsequent IOs each include the identifier for the new protocol endpoint and the additional identifier.

18. The method of claim 17, wherein each of the subsequent IOs is SCSI compliant and the identifier for the new protocol endpoint is a world wide name for a LUN and the additional identifier is a randomly generated number that is unique amongst logical storage volumes that are bound to the new protocol endpoint.

19. The method of claim 17, wherein each of the subsequent IOs is NFS compliant and the identifier for the new protocol endpoint is a mount point and an IP address for the storage system and the additional identifier is a file path to a file object in the NFS corresponding to the logical storage volume.

20. The method of claim 11, wherein the computer system has a virtual machine running therein and the logical storage volume corresponds to a virtual disk of the virtual machine.