Rapid Suspend/Resume for Virtual Machines via Resource Sharing

• US 20170192814A1
• Filed: 03/21/2017
• Published: 07/06/2017
• Est. Priority Date: 08/23/2014
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Beveridge, Daniel James, Snowdon, David
• Granted Patent: US 10,120,711 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 2009/45562 Creating, deleting, cloning virtual machine instances
• CPC Class Codes: G06F 2009/45575 Starting, stopping, suspending or resuming virtual machine instances
• CPC Class Codes: G06F 2009/45583 Memory management, e.g. access or allocation
• CPC Class Codes: G06F 2209/5011 Pool
• CPC Class Codes: G06F 9/4451 User profiles; Roaming
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects

Abstract

Examples quickly suspend and resume virtual desktops on demand or on schedule. Virtual desktops, or desktops as a service, are provided to users, where the virtual desktop is a forked VM, cloned VM, or otherwise at least a partial duplicate of an existing VM. The virtual desktop points to existing memory maintained by the existing VM, and the virtual desktop only writes to memory the pages that the virtual desktop creates or modifies.

Claims

1. A method comprising:

receiving a request to resume execution of a suspended child virtual machine (VM) on a host machine, the child VM being a child of a powered-on parent VM template;

retrieving memory blocks that are associated with the suspended child VM, but are not associated with the powered-on parent VM template; and

executing the suspended child VM to produce a powered-on child VM using memory blocks shared with the powered-on parent VM template and the retrieved memory blocks.

2. The method of claim 1, further comprising:

determining whether one or more of the retrieved memory blocks are compressed, and if so, decompressing the one or more memory blocks.

3. The method of claim 1, wherein the suspended child VM includes a log having pointers to a disk of the powered-on parent VM template as a base disk of the child VM.

4. The method of claim 3, wherein the log is non-persistent so that upon each execution of the suspended child VM, the powered-on child VM inherits a fresh copy of memory and device state from the powered-on parent VM template.

5. The method of claim 1, wherein the powered-on child VM is configured to use separate storage that is not related to the powered-on parent VM template.

6. The method of claim 1, wherein the suspended child VM includes configuration information indicating that the suspended child VM inherits memory and device state of the powered-on parent VM template upon execution.

7. The method of claim 1, further comprising:

receiving a request to suspend the powered-on child VM;

identifying memory blocks that are not shared with the powered-on parent VM template; and

transferring the identified memory blocks to storage.

8. The method of claim 1, wherein receiving the request comprises receiving the request from a user.

9. The method of claim 1, wherein the request is initiated at pre-programmed times in anticipation of requests from one or more users.

10. A system comprising:

a memory area associated with a computing device, said memory area storing a plurality of memory blocks associated with a plurality of virtual machines (VMs) and a processor programmed to perform operations comprising;

receiving a request to resume execution of a suspended child virtual machine (VM) on a host machine, the child VM being a child of a powered-on parent VM template;

retrieving memory blocks that are associated with the suspended child VM, but are not associated with the powered-on parent VM template; and

executing the suspended child VM to produce a powered-on child VM using memory blocks shared with the powered-on parent VM template and the retrieved memory blocks.

11. The system of claim 10 further configured to perform operations comprising:

determining whether one or more of the retrieved memory blocks are compressed, and if so, decompressing the one or more memory blocks.

12. The system of claim 10, wherein the suspended child VM includes a log having pointers to a disk of the powered-on parent VM template as a base disk of the child VM.

13. The system of claim 12, wherein the log is non-persistent so that upon each execution of the suspended child VM, the powered-on child VM inherits a fresh copy of memory and device state from the powered-on parent VM template.

14. The system of claim 10, wherein the powered-on child VM is configured to use separate storage that is not related to the powered-on parent VM template.

15. The system of claim 10, wherein the suspended child VM includes configuration information indicating that the suspended child VM inherits memory and device state of the powered-on parent VM template upon execution.

16. The system of claim 10, further configured to perform operations comprising:

receiving a request to suspend the powered-on child VM;

identifying memory blocks that are not shared with the powered-on parent VM template; and

transferring the identified memory blocks to storage.

17. The system of claim 10, wherein receiving the request comprises receiving the request from a user.

18. A method comprising:

receiving a request to suspend a child virtual machine (VM) executing on a host machine of a plurality of child VMs generated from a powered-on parent VM template resident on the same host, wherein each child VM shares a first plurality of memory blocks with the parent VM template, the shared memory blocks including boot pages and shared application pages, and wherein each child VM stores new writes to a separate plurality of non-shared memory blocks as copy on write versions of the corresponding shared pages and;

in response to the request to suspend the child VM;

identifying memory blocks of the child VM that are not shared with the parent VM template; and

transferring only the identified memory blocks of the child VM that are not shared with the parent VM template to a storage memory.

19. The method of claim 18, wherein the identified memory blocks of the child VM are transferred after a period of idle time.

20. The method of claim 18, wherein identifying memory blocks comprises comparing the memory blocks of the child VM to the memory blocks of the parent VM template.