VIRTUALIZED SERVER SYSTEMS AND METHODS INCLUDING LOAD BALANCING FOR VIRTUALIZED FILE SERVERS

• US 20180157521A1
• Filed: 12/01/2017
• Published: 06/07/2018
• Est. Priority Date: 12/02/2016
• Status: Active Grant

• Current Assignee: Nutanix Incorporated
• Original Assignee: Nutanix Incorporated
• Inventors: Arikatla, Durga Mahesh, Naik, Manoj, Gupta, Rashmi, Tammineedi, Veerraju, Huang, Yifeng, Rathi, Shyamsunder Prayagchand, Muthyala, Sravan Kumar, Ranasingh, Mausumi, Thummala, Hemanth Kumar, Gupta, Manish, Nair, Saji Kumar Vijaya Kumari Rajendran
• Granted Patent: US 10,824,455 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/188 Virtual file systems
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2009/45579 I/O management, e.g. providing access to device drivers or storage
• CPC Class Codes: G06F 2009/45583 Memory management, e.g. access or allocation
• CPC Class Codes: G06F 2009/45595 Network integration; Enabling network access in virtual machine instances
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects

Abstract

Examples described herein include virtualized file servers which may include load balancing. For example, a recommendation engine may estimate a load associated with each of a plurality of file server virtual machines in a cluster. Based on the load, the recommendation engine may recommend changing ownership of one or more volume groups, scaling up the virtualized file server, scaling down the virtualized file server, scaling in the virtualized file server, scaling out the virtualized file server, or combinations thereof.

Claims

1. A method comprising:

receive, at a file server virtual machine (an FSVM), input/output statistics for a plurality of volume groups in a storage pool managed by a virtualized file server, wherein the virtualized file server includes a plurality of file server virtual machines, including the file server virtual machine;

estimate a load of each of the plurality of file server virtual machines based on the input/output statistics; and

generate recommended changes to the virtualized file server based on the load of each of the plurality of file server virtual machines.

2. The method of claim 1, wherein estimate the load comprises identify a percentage of the input/output statistics handled by each of the plurality of file server virtual machines, and wherein estimate the load comprises estimating the load of each of the plurality of file server virtual machines at least in part by multiplying the percentage by a total amount of resources consumed by the virtualized file server.

3. The method of claim 2, wherein identify the percentage of input/output statistics handled by each of the plurality of file server virtual machines comprises summing input/output statistics for each volume group handled by a particular file server virtual machine of the plurality of file server virtual machines.

4. The method of claim 1, wherein the file server virtual machine is a master file server virtual machine of the plurality of file server virtual machines.

5. The method of claim 1, wherein the recommended changes comprising moving at least one of the volume groups from ownership by one of the file system virtual machines to ownership by another one of the file server virtual machines.

6. The method of claim 5, further comprising moving the at least one of the volume groups to ownership by the another one of the file server virtual machines at least in part by storing metadata indicative of ownership of the at least one of the volume groups by the another one of the file server virtual machines.

7. The method of claim 1, wherein the recommended changes comprise scaling out or scaling in the virtualized file server, and wherein the method further comprises adding or removing a number of file server virtual machines to or from the virtualized file server, and assigning ownership of selected volume groups owned by the number of file server virtual machines to the number of file server virtual machines or assigning ownership of selected volume groups owned by the number of file server virtual machines to others of the number of file server virtual machines.

8. The method of claim 1, wherein the recommended changes comprise scaling up or scaling down the virtualized file server, and wherein the method further comprises adding or subtracting memory, compute resources, or both to the plurality of file server virtual machines in the virtualized file server.

9. The method of claim 1, wherein the recommended changes comprise changing ownership of one or more of the volume groups, and wherein the method further comprises:

responsive to the recommended changes, reducing a cache time provided responsive to referral requests for the one or more of the volume groups.

10. The method of claim 9, wherein the method further comprises, responsive to an indication ownership of the one or more of the volume groups is complete, increasing the cache time provided responsive to subsequent referral requests for the one or more of the volume groups.

11. A system comprising:

a plurality of host machines implementing a virtualized file server, wherein each of the plurality of host machines comprises a hypervisor; and

a distributed file server comprising a plurality of file server virtual machines (a plurality of FSVMs) and a storage pool, wherein each of the plurality of FSVMs is running on one of the plurality of host machines, wherein the plurality of FSVMs are configured to conduct I/O transactions with the storage pool, the plurality of FSVMs comprises a master FSVM, the master FSVM configured to;

receive input/output statistics for a plurality of volume groups in the storage pool;

estimate a load of each of the plurality of file server virtual machines based on the input/output statistics; and

generate recommended changes to the virtualized file server based on the load of each of the plurality of file server virtual machines.

12. The system of claim 11, wherein the master FSVM is configured to estimate the load at least in part by identifying a percentage of the input/output statistics handled by each of the plurality of file server virtual machines, and multiplying the percentage by a total amount of resources consumed by the virtualized file server.

13. The system of claim 12, wherein the master FSVM is configured to identify the percentage of input/output statistics handled by each of the plurality of file server virtual machines at least in part by summing input/output statistics for each volume group handled by a particular file server virtual machine of the plurality of file server virtual machines.

14. The system of claim 11, wherein the system further comprises a controller virtual machine (a CVM) in communication with the master FSVM, and wherein the CVM is configured to provide the recommended changes to a user interface of an administrative system.

15. The system of claim 14, wherein the CVM is configured to implement the recommended changes responsive to a confirmation provided by the administrative system through the user interface.

16. The system of claim 14, wherein the recommended changes comprise moving at least one of the volume groups from ownership by one of the plurality of file server virtual machines to ownership by another one of the file server virtual machines, and wherein the CVM is further configured, responsive to the recommended changes, to reduce a cache time provided responsive to referral requests for the at least one of the volume groups.

17. The system of claim 16, further comprising a client configured to access the at least one of the volume groups at a location indicated by the referral received from the CVM.

18. The system of claim 16, wherein the CVM is further configured, responsive to an indication a change of ownership of the at least one of the volume groups is complete, increasing the cache time provided responsive to subsequent referral requests for the at least one of the volume groups.

19. The system of claim 14, wherein the recommended changes comprise scaling out or scaling in the virtualized file server, and wherein the CVM is further configure to add or remove a number of file server virtual machines to or from the virtualized file server, and assign ownership of selected volume groups owned by the plurality of file server virtual machines to the number of file server virtual machines or assign ownership of selected volume groups owned by the number of file server virtual machines to others of the plurality of file server virtual machines.

20. The system of claim 14, wherein the recommended changes comprise scaling up or scaling down the virtualized file server, and wherein the CVM is further configured to add or subtract memory, compute resources, or both to the plurality of file server virtual machines in the virtualized file server.