Method for dynamic load balancing on partitioned systems

• US 20080163239A1
• Filed: 12/29/2006
• Published: 07/03/2008
• Est. Priority Date: 12/29/2006
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Panesar, Kiran, Sugumar, Suresh
• Granted Patent: US 7,673,113 B2
• Time in Patent Office: Days
• US Class Current: 718/105
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2209/5022 Workload threshold
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5088 involving task migration

Abstract

Methods, systems and apparatuses to dynamically balance execution loads on a partitioned system among processor cores or among partitions.

Claims

1. A computing apparatus, comprising:

a plurality of partitions, wherein each partition includes one or more virtual machines to execute thereon;

a first computer instruction to determine a first load status of a first partition;

a second computer instruction to determine a second load status of a second partition; and

a third computer instruction to migrate one of the virtual machines between the first partition and the second partition when the first load status and the second load status match a pre-determined criteria.

2. The computing apparatus of claim 1, wherein the first and second partitions each comprise one or more processor cores.

3. The computing apparatus of claim 1, wherein the first partition comprises either a user partition or a system partition, and wherein the second partition comprises either a user partition or a system partition.

4. The computing apparatus of claim 1, wherein the pre-determined criteria and computer instructions are encoded in hardware, or software, or firmware, or processor core micro-code, or a combination thereof.

5. The computing apparatus of claim 1, wherein the third computer instruction to migrate the virtual machine based on a categorization attribute, in addition to the first and second load status matching the pre-determined criteria, and wherein the categorization attribute is selected from a group of categorization attributes consisting of access rights, priority, criticality, application type, operating system type, user type, VM processor core temperature, and requestor.

6. The computing apparatus of claim 1, further comprising:

a virtual machine monitor to generate the computer instructions for execution on the computing apparatus.

7. The computing apparatus of claim 1, further comprising:

a virtual machine monitor to modify a value of the predetermined criteria.

8. The computing apparatus of claim 1, further comprising:

a first virtual machine monitor to manage the virtual machines to execute on the first and second partitions.

9. The computing apparatus of claim 8, wherein the first virtual machine monitor to manage the one or more virtual machines to execute on the first partition, and further comprising:

a second virtual machine monitor to manage a second one or more virtual machines to execute on the second partition.

10. The computing apparatus of claim 9, further comprising:

a communication channel to connect the first virtual machine monitor with the second virtual machine monitor to exchange data corresponding with the first and second partitions; and

wherein the communication channel comprises either a shared memory component, or a direct memory access component, or both.

11. The computing apparatus of claim 1, further comprising:

a fourth computer instruction to quarantine a fault of any one virtual machine from any other virtual machine.

12. The computing apparatus of claim 1, further comprising:

a fifth computer instruction to delay the migration of the one virtual machine for a period of time.

13. A method of computing, comprising:

executing a plurality of partitions on a single host, wherein each partition includes one or more virtual machines executing thereon;

determining a first load status of a first partition;

determining a second load status of a second partition; and

migrating a virtual machine between the first partition and the second partition when the first load status and the second load status match a pre-determined criteria.

14. The method of computing of claim 13, wherein the first and second partitions each comprise one or more processor cores.

15. The method of computing of claim 13, wherein the first partition comprises either a user partition or a system partition, and wherein the second partition comprises either a user partition or a system partition.

16. The method of computing of claim 13, further comprising:

offloading work from one or more of the partitions to system resources, or user resources, or a combination thereof.

17. The method of computing of claim 13, further comprising:

dynamically modifying the frequency at which the first load status and second load status are determined based on a current load status of the first partition.

18. The method of computing of claim 13, wherein the pre-determined criteria comprises a lower threshold for the first partition and an upper threshold for the second partition, wherein the method further comprises:

executing a first test determining if the first load status of the first partition is below the lower threshold;

executing a second test determining if the second load status of the second partition is above the upper threshold; and

indicating the pre-determined criteria is matched when the first and second tests are true.

19. The method of computing of claim 18, wherein the pre-determined criteria further comprises a maximum available load value of the first partition and a minimum required load value of the second partition, wherein the method further comprises:

querying each of the virtual machines executing on the second partition for their respective minimum required load values;

comparing the minimum required load values received with the maximum available load value;

selecting the one virtual machine from the second partition with either the largest or the smallest minimum required load value that is less than or equal to the maximum available load value of the first partition; and

indicating the pre-determined criteria is matched to continue migrating the one selected virtual machine from the second partition to the first partition.

20. The method of computing of claim 13, wherein the pre-determined criteria includes a native task list and an upper threshold for the first partition, wherein the method further comprises:

executing a first test determining if the upper threshold has been exceeded;

executing a second test determining if the first partition is running a task not present on the native task list; and

indicating the pre-determined criteria is matched when the first and second tests are true.

21. The method of computing of claim 13, wherein migrating a virtual machine between the first partition and the second partition further comprises the virtual machine:

executing in a first state of pre-migration, wherein the virtual machine is executing as normal;

executing in a second state of reservation, wherein the virtual machine is executing on the first partition and corresponds to a newly initialized virtual machine on the second partition;

executing in a third state of pre-copy, wherein a plurality of memory pages of the virtual machine are copied from the virtual machine to the newly initialized virtual machine;

executing in a fourth state of stop-and-copy, wherein the virtual machine stops execution and any remaining memory pages of the plurality of memory pages corresponding with the virtual machine that are out of synchronization with the plurality of memory pages corresponding with the newly initialized virtual machine are re-copied to the newly initialized virtual machine;

executing in a fifth state of commitment, wherein the virtual machine corresponding to the first partition is eliminated and the newly initialized virtual machine is then referred to as the virtual machine; and

executing in a sixth state of activation, wherein the virtual machine connects to local devices and resumes normal operation by entering into a first state of pre-migration.

22. A computing apparatus, comprising:

a plurality of partitions, wherein each partition includes one or more virtual machines to execute thereon;

means to determine a first load status of a first partition;

means to determine a second load status of a second partition; and

means to migrate a virtual machine between the first partition and second partition if the first load status and the second load status match a pre-determined criteria.

23. The computing apparatus of claim 22, further comprising:

a means to balance a plurality of execution loads among the plurality of partitions.

24. The computing apparatus of claim 22, wherein the means to migrate a virtual machine further comprises:

a means to migrate a connected internet application executing within the virtual machine without disconnecting the connected internet application.

25. A system, comprising:

a plurality of partitions, wherein each partition includes one or more virtual machines to execute thereon;

a first computer instruction to determine a first load status of a first partition;

a second computer instruction to determine a second load status of a second partition; and

a third computer instruction to migrate one of the virtual machines between the first partition and the second partition if the first load status and the second load status match a pre-determined criteria; and

a communication interface to transmit a plurality of data between the system and a plurality of other systems connected with the communication interface.

26. The system of claim 25, further comprising:

a first virtual machine monitor to manage the one or more virtual machines to execute on the first partition; and

a second virtual machine monitor to manage the one or more virtual machines to execute on the second partition.

27. The system of claim 25, wherein the communication interface is selected from a group of communication interfaces consisting of Ethernet, coaxial, twisted pair, infrared, optical, Bluetooth, wireless, radio, InfiniBand, FCAL, serial, and satellite.

28. An article of manufacture, including a computer-readable medium having instructions to be executed by a machine, the machine having a plurality of partitions, wherein each partition includes one or more virtual machines to execute thereon, the instructions comprising:

a first instruction to determine a first load status of a first partition;

a second instruction to determine a second load status of a second partition; and

a third instruction to migrate one of the virtual machines between the first partition and the second partition when the first load status and the second load status match a pre-determined criteria.

29. The article of manufacture of claim 28, wherein the third computer instruction to migrate the virtual machine based on a categorization attribute, in addition to the first and second load status matching the pre-determined criteria, and wherein the categorization attribute is selected from a group of categorization attributes consisting of access rights, priority, criticality, application type, operating system type, user type, VM processor core temperature, and requestor.

30. The article of manufacture of claim 28, further comprising:

a fourth computer instruction to offload work from one or more of the partitions to system resources, or user resources, or a combination thereof.