METHODS AND APPARATUS TO GET FEEDBACK INFORMATION IN VIRTUAL ENVIRONMENT FOR SERVER LOAD BALANCING

• US 20100274890A1
• Filed: 04/28/2009
• Published: 10/28/2010
• Est. Priority Date: 04/28/2009
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Albert, Mark, Patel, Alpesh S., O'Rourke, Chris, Srinivasan, Udayakumar
• Granted Patent: US 9,600,332 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: G06F 9/4862 the task being a mobile agent, i.e. specifically designed to migrate
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 47/125 by balancing the load, e.g. traffic engineering

Abstract

Methods and apparatus for providing availability information of a virtual machine to a load balancer are disclosed. The availability information of the virtual machine may be normalized information from performance metrics of the virtual machine and performance metrics of the physical machine on which the virtual machine operates. The normalized availability of a virtual machine is provided by a feedback agent executing on the virtual machine. Alternatively, the normalized availability of a virtual machine is provided by a feedback agent executing on a hypervisor executing multiple virtual machines on a common set of physical computing hardware.

Claims

1. A computing system, comprising:

a processor;

a network interface;

a memory containing a hypervisor configured to execute a plurality of virtual machines on the computing system; and

a feedback agent configured to perform an operation to monitor an availability of at least a first virtual machine, of the plurality, the operation comprising;

monitoring one or more performance metrics of the first virtual machine,

monitoring one or more performance metrics of the computing system,

normalizing the one or more performance metrics of the first virtual machine based on the performance metrics of the computing system, and

sending, over the network interface, the normalized performance metrics of the virtual machine to a load balancer.

2. The computing system of claim 1, wherein the load balancer comprises a program executing on a second computing system and wherein the load balancer is configured to distribute computing tasks among the plurality of virtual machines based on the normalized performance metrics.

3. The computing system of claim 2, wherein the computing tasks comprise IP network service requests, and wherein the load balancer is configured to assign each request to a network connection on one of the virtual machines.

4. The computing system of claim 2, wherein load balancer is further configured to monitor a migration of one of the virtual machines from the computing system to a target computing system and to selectively drop network traffic addressed to connections on the virtual machine while the virtual machine is migrated to the target computing system.

5. The computing system of claim 1, wherein the feedback agent is a component of the hypervisor.

6. The computing system of claim 1, wherein the one or more performance metrics of the first virtual machine comprise at least one of a load on a virtual CPU and a load on a virtual network interface.

7. The computing system of claim 1, wherein the one or more performance metrics of the computing system comprise at least one of a load on the CPU and a load on the network interface, an indication of cache contention, and an indication of memory bus utilization.

8. The computing system of claim 1, wherein an instance of the feedback agent executes on each of the plurality of virtual machines.

9. The computing system of claim 8, wherein the instance of the feedback agent on one of the plurality of virtual machines is configured to obtain the monitoring performance metrics of the computing system by querying the hypervisor.

10. A computer-implemented method for providing feedback information for load balancing applications executing on a plurality of virtual machines, the method comprising:

monitoring, by a feedback agent, one or more performance metrics of a first virtual machine, of the plurality;

monitoring one or more performance metrics of a first computing system hosting the first virtual machine;

normalizing the one or more performance metrics of the first virtual machine based on the performance metrics of the first computing system; and

sending, over a network interface, the normalized performance metrics of the virtual machine to a load balancer application.

11. The computer-implemented method of claim 10, wherein the load balancer application comprises a program executing on a second computing system and wherein the load balancer application is configured to distribute computing tasks among the plurality of virtual machines based on the normalized performance metrics.

12. The computer-implemented method of claim 11, wherein the computing tasks comprise an IP network service requests, and wherein the load balancer is configured to:

assign each request to a network connection on one of the virtual machines;

monitor a migration of one of the virtual machines from the first computing system to a target computing system; and

selectively drop network traffic addressed to connections on the virtual machine being migrated during the migration.

13. The computer-implemented method of claim 10, wherein the feedback agent is a component of a hypervisor executing the first computing system.

14. The computer-implemented method of claim 10, wherein the one or more performance metrics of the first virtual machine comprise at least one of a load on a virtual CPU and a load on a virtual network interface.

15. The computer-implemented method of claim 10, wherein the one or more performance metrics of the first computing system comprise at least one of a load on a CPU and a load on the network interface.

16. The computer-implemented method of claim 10, wherein an instance of the feedback agent executes on each of the plurality of virtual machines.

17. The computer-implemented method of claim 15, wherein the instance of the feedback agent on one of the plurality of virtual machines is configured to obtain the monitoring performance metrics of the computing system by querying the hypervisor.

18. A computer-readable storage medium containing a feedback agent which, when executed on a processor, performs an operation for providing feedback information to load balance applications executing on a plurality of virtual machines, the operation comprising:

monitoring, by the feedback agent, one or more performance metrics of a first virtual machine, of the plurality;

monitoring one or more performance metrics of a first computing system hosting the first virtual machine;

normalizing the one or more performance metrics of the first virtual machine based on the performance metrics of the first computing system; and

sending, over a network interface, the normalized performance metrics of the virtual machine to a load balancer application.

19. The computer-readable storage medium of claim 18, wherein the load balancer application comprises a program executing on a second computing system and wherein the load balancer application is configured to distribute computing tasks among the plurality of virtual machines based on the normalized performance metrics.

20. The computer-readable storage medium of claim 19, wherein the computing tasks comprise IP network service requests, and wherein the load balancer is configured to:

assign each request to a network connection on one of the virtual machines;

monitor a migration of one of the virtual machines from the first computing system to a target computing system; and

selectively drop network traffic addressed to connections on the virtual machine being migrated during the migration.

21. The computer-readable storage medium of claim 18, wherein the feedback agent is a component of a hypervisor executing the first computing system.

22. The computer-readable storage medium of claim 18, wherein the one or more performance metrics of the first virtual machine comprise at least one of a load on a virtual CPU and a load on a virtual network interface.

23. The computer-readable storage medium of claim 18, wherein the one or more performance metrics of the first computing system comprise at least one of a load on a CPU, a load on the network interface, an indication of cache contention, and an indication of memory bus utilization.

24. The computer-readable storage medium of claim 18, wherein an instance of the feedback agent executes on each of the plurality of virtual machines.

25. The computer-readable storage medium of claim 24, wherein the instance of the feedback agent on one of the plurality of virtual machines is configured to obtain the monitoring performance metrics of the computing system by querying the hypervisor.