Virtual machine switching based on measured network delay

• US 9,075,648 B2
• Filed: 04/27/2012
• Issued: 07/07/2015
• Est. Priority Date: 04/27/2012
• Status: Expired due to Fees

• Current Assignee: Invincible IP LLC (IP Edge LLC)
• Original Assignee: Empire Technology Development LLC (Allied Inventors Management, LLC)
• Inventors: Kruglick, Ezekiel
• Primary Examiner(s): Puente, Emerson
• Assistant Examiner(s): Huaracha, Willy W
• Application Number:
  US13/582,912
• Publication Number:
  US 20130290956A1
• Time in Patent Office: 1,166 Days
• Field of Search: 718/1
• US Class Current: 1/1
• CPC Class Codes: G06F 2009/45575 Starting, stopping, suspending or resuming virtual machine instances
• 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
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 47/283 in response to processing delays, e.g. caused by jitter or round trip time [RTT]
• CPC Class Codes: H04L 67/142 Managing session states for stateless protocols; Signalling session states; State transitions; Keeping-state mechanisms

Abstract

Technologies related to virtual machine switching based on measured network delay are generally described. A network delay aware Virtual Machine (VM) may be configured to adapt a Virtual Machine Monitor (VMM) to delay switching back to the network delay aware VM by a delay amount determined using a measured network delay. The measured network delay may comprise a delay between sending a network communication and receiving a network response. By delaying switching back to the network delay aware VM, additional processing resources are freed for other VMs managed by the VMM, thereby increasing efficiency of computing devices including network delay aware VMs, and correspondingly increasing efficiency of data centers including such computing devices.

Claims

1. A server comprising:

a processor;

a network delay aware Virtual Machine Manager (VMM) configured to manage execution of a plurality of Virtual Machines (VMs) by the processor, wherein managing execution of the plurality of VMs comprises switching the processor between execution of different VMs,

wherein the network delay aware VMM comprises an Application Program Interface (API) configured to receive a measured network delay from at least one network delay aware VM managed by the network delay aware VMM, and

wherein the network delay aware VMM is configured to adapt a scheduler to delay switching the processor back to execution of the at least one network delay aware VM by approximately the measured network delay received at the API from the at least one network delay aware VM; and

the at least one network delay aware VM, comprising;

a delay measurement module configured to measure network delay between sending a network communication by the at least one network delay aware VM and receiving a network response by the at least one network delay aware VM, to thereby determine the measured network delay; and

a timing signaler module configured to provide the measured network delay to the API at the network delay aware VMM in order to adapt the network delay aware VMM to delay switching the processor back to execution of the at least one network delay aware VM by approximately the measured network delay.

2. The server of claim 1, wherein the measured network delay comprises a network communications delay plus an action delay, wherein the action delay comprises delay associated with a user action to trigger the network response.

3. The server of claim 1, wherein the at least one network delay aware VM is configured with one or more game modules.

4. The server of claim 1, wherein the delay measurement module is configured to measure network delay at least in part by requesting delay data from a Session Border Controller (SBC).

5. The server of claim 1, wherein the server is configured to manage execution of other VMs than the at least one network delay aware VM, using a fair-sharing technique to switch the processor between execution of the other VMs.

24. The server of claim 1, wherein the delay measurement module is configured to measure network delay at least in part by calculating a difference between an outgoing network communication time stamp and an incoming network response time stamp.

25. The server of claim 1, wherein the delay measurement module is configured to retrieve communication history data from a communication device, the communication history data comprising times of network communications and network responses.

26. The server of claim 1, wherein the delay measurement module is configured to measure network delay at least in part by starting a clock when the network communication is sent and stopping the clock when the network response is received.

27. The server of claim 1, wherein:

the delay measurement module is configured to measure network delay for a subset of cloud client devices associated with relatively slow response times that are longer than a predetermined threshold value; and

the at least one network delay aware VM is configured to adapt the network delay aware VMM to delay switching the processor back to execution of the at least one network delay aware VM in connection with communications between the at least one network delay aware VM and the subset of cloud client devices.

6. A non-transitory computer readable storage medium having computer executable instructions executable by a processor, the instructions that, when executed by the processor, cause the processor to:

measure, by a network delay aware Virtual Machine (VM), a network delay between sending a network communication by the network delay aware VM and receiving a network response by the network delay aware VM, to thereby determine a measured network delay; and

provide the measured network delay to an API at a network delay aware Virtual Machine Manager (VMM) in order to adapt, by the network delay aware VM, the network delay aware VMM to delay switching the processor back to execution of the network delay aware VM by approximately the measured network delay, wherein the network delay aware VMM is configured to manage execution of a plurality of VMs by the processor, wherein managing execution of the plurality of VMs comprises switching the processor between execution of different VMs, and wherein the network delay aware VMM is configured to adapt a scheduler to delay switching the processor back to execution of the network delay aware VM by approximately the measured network delay received at the API from the network delay aware VM.

7. The non-transitory computer readable storage medium of claim 6, wherein the measured network delay comprises a network communications delay plus an action delay, wherein the action delay comprises delay associated with a user action to trigger the network response.

8. The non-transitory computer readable storage medium of claim 6, further comprising instructions that cause the processor to execute a game in the network delay aware VM, and wherein the network communication and network response are sent and received in connection with the game.

9. The non-transitory computer readable storage medium of claim 6, wherein the instructions that cause the processor to measure network delay are configured to request delay data from a Session Border Controller (SBC).

10. A method comprising:

measuring, by a network delay aware Virtual Machine (VM), a network delay between sending a network communication by the network delay aware VM and receiving a network response by the network delay aware VM, to thereby determine a measured network delay; and

providing, by the network delay aware VM, the measured network delay to an API at a network delay aware Virtual Machine Manager (VMM) in order to adapt the network delay aware VMM to delay switching the processor back to execution of the network delay aware VM by approximately the measured network delay, wherein the network delay aware VMM is configured to manage execution of a plurality of VMs by the processor, wherein managing execution of the plurality of VMs comprises switching the processor between execution of different VMs, and wherein the network delay aware VMM is configured to adapt a scheduler to delay switching the processor back to execution of the network delay aware VM by approximately the measured network delay received at the API from the network delay aware VM.

11. The method of claim 10, wherein the measured network delay comprises a network communications delay plus an action delay, wherein the action delay comprises delay associated with a user action to trigger the network response.

12. The method of claim 10, further comprising executing a game in the network delay aware VM, and wherein the network communication and network response are sent and received in connection with the game.

13. The method of claim 10, wherein the instructions that cause the processor to measure network delay are configured to request delay data from a Session Border Controller (SBC).

14. A non-transitory computer readable storage medium having computer executable instructions executable by a processor, the instructions that, when executed by the processor, cause the processor to:

manage, by a network delay aware Virtual Machine Manager (VMM), execution of a plurality of Virtual Machines (VMs) by the processor, wherein managing execution of the plurality of VMs comprises switching the processor between execution of different VMs;

provide an Application Program Interface (API) by the network delay aware VMM;

receive by the API a network delay input comprising a network delay measured between sending a network communication by a network delay aware VM and receiving a network response by the network delay aware VM; and

in response to receiving the network delay input by the API, adapt a scheduler to delay switching the processor back to execution of the network delay aware VM by approximately a delay amount determined using the network delay input.

15. The non-transitory computer readable storage medium of claim 14, wherein the network delay measured between sending a network communication by the network delay aware VM and receiving a network response by the network delay aware VM comprises a network communications delay plus an action delay, wherein the action delay comprises delay associated with a user action to trigger the network response.

16. The non-transitory computer readable storage medium of claim 14, further comprising instructions that cause the processor to co-locate VMs other than the network delay aware VM, which are associated with a same data center customer, on a server comprising the network delay aware VMM and network delay aware VM so that the co-located VMs benefit from processor time freed by delay in switching back to execution of the network delay aware VM.

17. The non-transitory computer readable storage medium of claim 14, further comprising instructions that cause the processor to manage execution of other VMs, other than the network delay aware VM, using a fair-sharing technique to time switching the processor between execution of the other VMs.

18. The non-transitory computer readable storage medium of claim 14, further comprising instructions that cause the processor to:

receive by the API a notification from the network delay aware VM;

retrieve, in response to the notification, a network delay input from a Session Border Controller (SBC), wherein the network delay input from the SBC comprises network delay measured at the SBC and associated with network communications and network responses of the of the network delay aware VM; and

adapt the scheduler to delay switching the processor back to execution of the network delay aware VM by approximately a delay amount determined using the network delay input from the SBC.

19. A method comprising:

managing, by a network delay aware Virtual Machine Manager (VMM), execution of a plurality of Virtual Machines (VMs) by a processor, wherein managing execution of the plurality of VMs comprises switching the processor between execution of different VMs;

providing an Application Program Interface (API) by the network delay aware VMM;

receiving by the API a network delay aware Virtual Machine (VM) network delay input comprising a network delay measured between sending a network communication by the network delay aware VM and receiving a network response by the network delay aware VM; and

in response to receiving the network delay input by the API, adapting a scheduler to delay switching the processor back to execution of the network delay aware VM by approximately a delay amount determined using the network delay input.

20. The method of claim 19, wherein the network delay measured between sending a network communication by the network delay aware VM and receiving a network response by the network delay aware VM comprises a network communications delay plus an action delay, wherein the action delay comprises delay associated with a user action to trigger the network response.

21. The method of claim 19, further comprising co-locating other VMs, other than the network delay aware VM, which other VMs are associated with a same data center customer, on a server comprising the network delay aware VMM and network delay aware VM so that the other VMs benefit from processor time freed by delay in switching back to execution of the network delay aware VM.

22. The method of claim 21, further comprising managing execution of the other VMs using a fair-sharing technique to switch the processor between execution of the other VMs.

23. The method of claim 19, further comprising:

receiving by the API a notification from the network delay aware VM;

retrieving, in response to the notification, a network delay input from a Session Border Controller (SBC), wherein the network delay input from the SBC comprises network delay measured at the SBC and associated with network communications and network responses of the of the network delay aware VM; and

adapting the scheduler to delay switching the processor back to execution of the network delay aware VM by approximately a delay amount determined using the network delay input from the SBC.