Virtual machine monitor configured to support latency sensitive virtual machines

Patent metadata

Field	Value
Publication number	US-10073711-B2
Application number	US-201615097035-A
Country	US
Kind code	B2
Filing date	Apr 12, 2016
Priority date	Aug 26, 2013
Publication date	Sep 11, 2018
Grant date	Sep 11, 2018

Abstract

A host computer has a virtualization software that supports execution of a plurality of virtual machines, where the virtualization software includes a virtual machine monitor for each of the virtual machines, and where each virtual machine monitor emulates a virtual central processing unit (CPU) for a corresponding virtual machine. A virtual machine monitor halts execution of a virtual CPU of a virtual machine by receiving a first halt instruction from a corresponding virtual machine and determining whether the virtual machine is latency sensitive. If the virtual machine is latency sensitive, then a second halt instruction is issued from the virtual machine monitor to halt a physical CPU on which the virtual CPU executes. If the virtual machine is not latency sensitive, then a system call to a kernel executing on the host computer is executed to indicate to the kernel that the virtual CPU is in an idle state.

First claim

We claim:

1. In a host computer having a virtualization software that supports execution of a plurality of virtual machines, the virtualization software including a virtual machine monitor for each of the virtual machines, wherein each virtual machine monitor emulates a virtual central processing unit (CPU) for a corresponding virtual machine, a method of halting execution of a virtual CPU of a virtual machine, the method comprising:
receiving a first halt instruction at a virtual machine monitor from a corresponding virtual machine;
determining by the virtual machine monitor that the virtual machine is latency sensitive by reading a latency sensitivity indicator for the virtual machine and a CPU entitlement value for the virtual machine, and determining that the latency sensitivity indicator is a predetermined value and that the CPU entitlement value is a maximum value; and
upon determining that the virtual machine is latency sensitive, issuing from the virtual machine monitor a second halt instruction to halt a physical CPU on which the virtual CPU executes.

2. The method of claim 1 , wherein the first halt instruction is issued by a guest operating system executing in the virtual machine.

3. The method of claim 1 , wherein the virtual CPU executes in an exclusive mode on the physical CPU.

4. The method of claim 1 , wherein said issuing from the virtual machine monitor the second halt instruction includes making a system call to a kernel CPU scheduler.

5. The method of claim 4 , wherein the kernel CPU scheduler, in response to the system call, deschedules the virtual CPU.

6. The method of claim 5 , wherein, in response to the system call, the kernel CPU scheduler deallocates the physical CPU from the virtual CPU and allocates the physical CPU to other tasks.

7. A non-transitory computer-readable medium comprising instructions executable by a host computer, the host computer having a virtualization software that supports execution of a plurality of virtual machines, the virtualization software including a virtual machine monitor for each of the virtual machines, wherein each virtual machine monitor emulates a virtual central processing unit (CPU) for a corresponding virtual machine, where the instructions, when executed, cause the host computer to perform method of halting execution of a virtual CPU of a virtual machine, the method comprising:
receiving a first halt instruction at a virtual machine monitor from a corresponding virtual machine;
determining by the virtual machine monitor that the virtual machine is latency sensitive by reading a latency sensitivity indicator for the virtual machine and a CPU entitlement value for the virtual machine, and determining that the latency sensitivity indicator is a predetermined value and that the CPU entitlement value is a maximum value; and
upon determining that the virtual machine is latency sensitive, issuing from the virtual machine monitor a second halt instruction to halt a physical CPU on which the virtual CPU executes.

8. The computer-readable medium of claim 7 , wherein the first halt instruction is issued by a guest operating system executing in the virtual machine.

9. The computer-readable medium of claim 7 , wherein the virtual CPU executes in an exclusive mode on the physical CPU.

10. The computer-readable medium of claim 7 , wherein said issuing from the virtual machine monitor the second halt instruction includes making a system call to a kernel CPU scheduler.

11. The computer-readable medium of claim 10 , wherein the kernel CPU scheduler, in response to the system call, deschedules the virtual CPU.

12. The computer-readable medium of claim 11 , wherein, in response to the system call, the kernel CPU scheduler deallocates the physical CPU from the virtual CPU and allocates the physical CPU to other tasks.

13. A virtualized computing system, comprising:
a host computer having a virtualization software that supports execution of a plurality of virtual machines;
a kernel scheduler; and
a virtual machine monitor for each of the virtual machines, wherein each virtual machine monitor emulates a virtual central processing unit (CPU) for a corresponding virtual machine, wherein each virtual machine monitor is configured to perform a method of halting execution of a virtual CPU of a virtual machine, the method comprising:
receiving a first halt instruction at a virtual machine monitor from a corresponding virtual machine;
determining by the virtual machine monitor that the virtual machine is latency sensitive by reading a latency sensitivity indicator for the virtual machine and a CPU entitlement value for the virtual machine, and determining that the latency sensitivity indicator is a predetermined value and that the CPU entitlement value is a maximum value; and
upon determining that the virtual machine is latency sensitive, issuing from the virtual machine monitor a second halt instruction to halt a physical CPU on which the virtual CPU executes.

14. The system of claim 13 , wherein the first halt instruction is issued by a guest operating system executing in the virtual machine.

15. The system of claim 13 , wherein the virtual CPU executes in an exclusive mode on the physical CPU.

16. The system of claim 13 , wherein said issuing from the virtual machine monitor the second halt instruction includes making a system call to a kernel CPU scheduler.

17. The system of claim 16 , wherein the kernel CPU scheduler, in response to the system call, deschedules the virtual CPU, deallocates the physical CPU from the virtual CPU, and allocates the physical CPU to other tasks.

Assignees

• Vmware Inc
• Wmware Inc

Inventors

• Zheng Haoqiang
• Singaravelu Lenin
• Agarwal Shilpi
• Hecht Daniel Michael
• Smith Garrett

Classifications

• G06F2009/4557

  Distribution of virtual machine instances; Migration and load balancing · CPC title

• G06F9/50

  Allocation of resources, e.g. of the central processing unit [CPU] · CPC title

• H04L47/801

  Real time traffic · CPC title

• G06F9/5033

  considering data affinity · CPC title

• G06F9/45533

  Hypervisors; Virtual machine monitors · CPC title