Method and system for synchronizing fault-tolerant virtual machines and adjusting CPU resource limit based on execution latency

• US 8,650,564 B2
• Filed: 10/19/2010
• Issued: 02/11/2014
• Est. Priority Date: 10/19/2010
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Huang, Lan, Scales, Daniel J.
• Primary Examiner(s): An, Meng
• Assistant Examiner(s): TEETS, BRADLEY A
• Application Number:
  US12/907,687
• Publication Number:
  US 20120096458A1
• Time in Patent Office: 1,211 Days
• Field of Search: None
• US Class Current: 718/1
• CPC Class Codes: G06F 11/1484 involving virtual machines
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455

Abstract

Primary and secondary virtual machines each executing on a separate physical host and forming a fault-tolerant pair synchronize their execution with each other as closely as possible such that the secondary can take over execution of the primary with little or no disruption of service and no loss of data. To provide fast takeover, the execution latency between the two virtual machines is kept to a minimum by incrementally adjusting the CPU allocation to the primary virtual machine. The CPU allocation to the primary virtual machine decreases when the execution latency is large and increases when the execution latency is small. In order to maximize the performance of the primary virtual machine, the system seeks to maximize the CPU resource limit for the primary virtual machine for as long as possible while minimizing the execution latency.

Claims

1. A method for synchronizing execution of a first virtual machine and a second virtual machine which are organized as a fault-tolerant pair of virtual machines, wherein the first virtual machine is a primary virtual machine and executes within a first computer system and the second virtual machine is enabled to track the state of the primary virtual machine and executes within a second computer system, the method comprising:

determining an execution latency between the first virtual machine and the second virtual machine, wherein the execution latency represents the difference in time between a first time at which an event is executed at the first virtual machine and a second time at which the same event is executed at the second virtual machine; and

adjusting a CPU resource limit in the first virtual machine in proportion to the determined execution latency, wherein the CPU resource limit is an allocation of CPU processing capability of the first computer system.

2. The method of claim 1, wherein determining the execution latency comprises:

sampling an event at the first virtual machine and recording a first time at which the event was executed at the first virtual machine;

receiving the same event at the second virtual machine;

executing the same event at the second virtual machine and recording a second time at which the same event was executed at the second virtual machine; and

computing the execution latency by subtracting the first time from the second time.

3. The method of claim 1, wherein adjusting the CPU resource limit comprises:

throttling the CPU resource limit based on the determined execution latency and other pre-determined threshold parameters; and

stabilizing the CPU resource limit at a maximum CPU resource limit value.

4. The method of claim 3, wherein throttling the CPU resource limit comprises:

receiving the determined execution latency; and

if the determined execution latency exceeds a pre-defined threshold, reducing the CPU resource limit of the first virtual machine in proportion to the determined execution latency.

5. The method of claim 3, wherein stabilizing the CPU resource limit comprises:

setting the CPU resource limit to either the original maximum CPU resource limit or a fixed percentage of a newly computed CPU resource limit if the determined execution latency exceeds a pre-defined threshold.

6. The method of claim 1, wherein the primary virtual machine is active and interacts with the external world and the secondary virtual machine is passive and invisible to the external world as the secondary virtual machine tracks the state of the primary virtual machine.

7. The method of claim 1, wherein the CPU resource limit is expressed as a percentage of the CPU or the CPU speed.

8. A non-transitory computer-readable storage medium for storing instructions that, when executed by a computer system causes the computer system to perform a method for synchronizing execution of the first virtual machine and a second virtual machine which are organized as a fault-tolerant pair of virtual machines, wherein the first virtual machine is a primary virtual machine and executes within a first computer system and the second virtual machine is enabled to track the state of the primary virtual machine and executes within a second computer system, the method comprising:

determining an execution latency between the first virtual machine and the second virtual machine, wherein the execution latency represents the difference in time between a first time at which an event is executed at the first virtual machine and a second time at which the same event is executed at the second virtual machine; and

adjusting a CPU resource limit in the first virtual machine in proportion to the determined execution latency.

9. The non-transitory computer-readable storage medium of claim 8, wherein determining the execution latency comprises sampling an event at the first virtual machine and recording a first time at which the event was executed at the first virtual machine;

receiving the same event at the second virtual machine;

executing the same event at the second virtual machine and recording a second time at which the same event was executed at the second virtual machine; and

computing the execution latency by subtracting the first time from the second time.

10. The non-transitory computer-readable storage medium of claim 8, wherein adjusting the CPU resource limit comprises:

throttling the CPU resource limit based on the determined execution latency and other pre-determined threshold parameters; and

stabilizing the CPU resource limit at a maximum CPU resource limit value.

11. The non-transitory computer-readable storage medium of claim 10, wherein throttling the CPU resource limit comprises:

receiving the determined execution latency; and

if the determined execution latency exceeds a pre-defined threshold, reducing the CPU resource limit of the first virtual machine in proportion to the determined execution latency.

12. The non-transitory computer-readable storage medium of claim 10, wherein stabilizing the CPU resource limit comprises setting the CPU resource limit to either the original maximum CPU resource limit or a fixed percentage of a newly computed CPU resource limit if the determined execution latency exceeds a pre-defined threshold.

13. The non-transitory computer-readable storage medium of claim 8, wherein the CPU resource limit is expressed as a percentage of the CPU or the CPU speed.

14. A computer system that synchronizes execution of a first virtual machine and a second virtual machine which are organized as a fault-tolerant pair of virtual machines, wherein the first virtual machine is a primary virtual machine and executes within a first computer system and the second virtual machine is enabled to track the state of the primary virtual machine and executes within a second computer system, the computer system comprising:

a processing unit; and

a memory,

wherein the processing unit is configured to determine an execution latency between the first virtual machine and the second virtual machine, wherein the execution latency represents the difference in time between a first time at which an event is executed at the first virtual machine and a second time at which the same event is executed at the second virtual machine,

and adjust a CPU resource limit in the first virtual machine in proportion to the determined execution latency.

15. The computer system of claim 14, wherein the processing unit is further configured to determine the execution latency by:

sampling an event at the first virtual machine and recording a first time at which the event was executed at the first virtual machine;

receiving the same event at the second virtual machine;

executing the same event at the second virtual machine and recording a second time at which the same event was executed at the second virtual machine; and

computing the execution latency by subtracting the first time from the second time.

16. The computer system of claim 14, wherein the processing unit is further configured to adjust the CPU resource limit by:

throttling the CPU resource limit based on the determined execution latency and other pre-determined threshold parameters; and

stabilizing the CPU resource limit at a maximum CPU resource limit value.

17. The computer system of claim 16, wherein throttling the CPU resource limit comprises:

receiving the determined execution latency; and

if the determined execution latency exceeds a pre-defined threshold, reducing the CPU resource limit of the first virtual machine in proportion to the determined execution latency.

18. The computer system of claim 16, wherein stabilizing the CPU resource limit comprises setting the CPU resource limit to either the original maximum CPU resource limit or a fixed percentage of a newly computed CPU resource limit if the determined execution latency exceeds a pre-defined threshold.

19. The computer system of claim 14, wherein the primary virtual machine is active and interacts with the external world and the secondary virtual machine is passive and invisible to the external world as the secondary virtual machine tracks the state of the primary virtual machine.

20. The computer system of claim 14, wherein the CPU resource limit is expressed as a percentage of the CPU or the CPU speed.