Server load balancer scaling for virtual servers

• US 8,949,410 B2
• Filed: 09/10/2010
• Issued: 02/03/2015
• Est. Priority Date: 09/10/2010
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Patel, Alpesh, O'Rourke, Christopher
• Primary Examiner(s): Mejia, Anthony
• Application Number:
  US12/879,203
• Publication Number:
  US 20120066371A1
• Time in Patent Office: 1,607 Days
• Field of Search: 709/200, 709/224
• US Class Current: 709/224
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests

Abstract

Techniques are provided herein for receiving information at a device in a network indicating a load level for one or more server load balancers that are configured to manage network traffic load for a plurality of servers. The information represents an aggregate load across the plurality of servers. A determination is made as to whether the load level for one or more of the server load balancers exceeds a predetermined threshold. In response to determining that the load level for one or more of the server load balancers exceeds the predetermined threshold, an additional load balancer is activated that is configured to manage network traffic load for the plurality of servers. In response to determining that the load level for one or more of the server load balancers does not exceed the predetermined threshold, an additional load balancer is deactivated that was configured to manage network traffic load for the plurality of servers.

Claims

1. A method comprising:

at a device in a network that comprises one or more server load balancers configured to assign received network traffic to one of a plurality of servers based on the load of each of the plurality of servers, receiving information from the one or more server load balancers indicating a load level for one or more server load balancers and an aggregate load across the plurality of servers;

determining when the load level for at least one of the one or more of the server load balancers exceeds a predetermined threshold;

in response to determining that the load level for at least one of the one or more of the server load balancers exceeds the predetermined threshold, activating an additional server load balancer configured to assign received network traffic to one of the plurality of servers based on the load of each of the plurality of servers; and

dynamically allocating the received network traffic to the one or more server load balancers and the additional server load balancer, by performing dynamic modification of one or more Internet Protocol (IP) addresses of one or more respective server load balancers of the one or more server load balancers and the additional server load balancer,

wherein the device is an upstream device relative to the one or more server load balancers and a direction of network traffic.

2. The method of claim 1, further comprising:

in response to determining that the load level for at least one of the one or more of the server load balancers does not exceed a predetermined threshold, deactivating an additional server load balancer that was configured to manage network traffic load for the plurality of servers.

3. The method of claim 1, wherein activating comprises activating the additional server load balancer on a server load balancer device that was not previously configured to manage network traffic load for the plurality of servers.

4. The method of claim 1, wherein the device in the network is at least one of a network management device, or an upstream network distribution device.

5. The method of claim 1, wherein activating comprises activating the additional server load balancer by configuring the additional load balancer to perform route injection configured to advertise the additional server load balancer'"'"'s ability to take on load.

6. The method of claim 1, wherein activating comprises activating the additional server load balancer according to a priority assigned to two or more geographic sites each comprising a subset of the plurality of servers.

7. The method of claim 6, wherein one of the two or more geographic sites comprises a subset of the plurality servers that are not configured to serve a particular virtual service and further comprising activating an additional server at the one geographic site to handle additional traffic for the particular virtual service.

8. The method of claim 1, wherein activating comprises activating a virtual server load balancer.

9. The method of claim 8, wherein activating comprises instantiating a virtual machine comprising the virtual server load balancer.

10. The method of claim 9, wherein instantiating comprises instantiating the virtual server load balancer on an active server load balancer device configured to manage network traffic load for the plurality of servers.

11. An apparatus comprising:

one or more network interfaces configured to communicate over a network;

a processor coupled to the one or more network interfaces and configured to;

receive information over the network from one or more server load balancers configured to assign received network traffic to one of a plurality of servers based on the load of each of the plurality of servers, wherein the information indicates a load level for the one or more server load balancers and an aggregate load across the plurality of servers;

determine when the load level for at least one of the one or more of the server load balancers exceeds a predetermined threshold;

in response to determining that the load level for at least one of the one or more of the server load balancers exceeds the predetermined threshold, activate an additional server load balancer configured to assign received network traffic to one of the plurality of servers based on the load of each of the plurality of servers; and

dynamically allocate the received network traffic to the one or more server load balancers and the additional server load balancer, by performing dynamic modification of one or more Internet Protocol (IP) addresses of one or more respective server load balancers of the one or more server load balancers and the additional server load balancer,

wherein the apparatus is an upstream device relative to the one or more server load balancers and a direction of network traffic.

12. The apparatus of claim 11, wherein the processor is further configured to:

deactivate an additional server load balancer that was configured to manage network traffic load for the plurality of servers in response to determining that the load level for at least one of the one or more of the server load balancers does not exceed a predetermined threshold.

13. The apparatus of claim 11, wherein the processor is further configured to activate the additional server load balancer according to a priority assigned to two or more geographic sites each comprising a subset of the plurality of servers.

14. The apparatus of claim 11, wherein the processor is further configured to activate the additional server load balancer by instantiating a virtual machine comprising the virtual server load balancer.

15. A non-transitory computer readable medium storing instructions that, when executed by a processor, cause the processor to:

receive information from one or more server load balancers configured to assign received network traffic to one of a plurality of servers based on the load of each of the plurality of servers, wherein the information indicates a load level for the one or more server load balancers and an aggregate load across the plurality of servers;

determine when the load level for at least one of the one or more of the server load balancers exceeds a predetermined threshold;

in response to determining that the load level for at least one of the one or more of the server load balancers exceeds the predetermined threshold, activate an additional server load balancer configured to assign received network traffic to one of the plurality of servers based on the load of each of the plurality of servers; and

dynamically allocate the received network traffic to the one or more server load balancers and the additional server load balancer, by performing dynamic modification of one or more Internet Protocol (IP) addresses of one or more respective server load balancers of the one or more server load balancers and the additional server load balancer,

wherein the computer readable medium and processor are incorporated in a device that is upstream relative to the one or more server load balancers and a direction of network traffic.

16. The non-transitory computer readable medium of claim 15, further comprising instructions that, when executed by the processor, cause the processor to deactivate an additional server load balancer that was configured to manage network traffic load for the plurality of servers in response to determining that the load level for at least one of the one or more of the server load balancers does not exceed a predetermined threshold.

17. The non-transitory computer readable medium of claim 15, further comprising instructions that, when executed by the processor, cause the processor to activate the additional server load balancer according to a priority assigned to two or more geographic sites each comprising a subset of the plurality of servers.

18. The non-transitory computer readable medium of claim 15, further comprising instructions that, when executed by the processor, cause the processor to activate the additional server load balancer by instantiating a virtual machine for a virtual server load balancer.

19. A method comprising:

at a device in a network, generating information representing a load level for one or more virtual server load balancers operating on the device that are configured to assign received network traffic to one of a plurality of servers based on the load of each of the plurality of servers, wherein the information indicates a load level for the one or more virtual server load balancers and an aggregate load across the plurality of servers;

determining when the load level for at least one of the one or more of the virtual server load balancers exceeds a predetermined threshold;

in response to determining that the load level for one or more of the virtual server load balancers exceeds the predetermined threshold, instantiating an additional virtual server load balancer configured to assign received network traffic to one of the plurality of servers based on the load of each of the plurality of servers; and

dynamically allocating the received network traffic to the one or more virtual server load balancers and the additional virtual server load balancer, by performing dynamic modification of one or more Internet Protocol (IP) addresses of one or more respective virtual server load balancers of the one or more virtual server load balancers and the additional server load balancer.

20. The method of claim 19, further comprising collaborating between the one or more virtual server load balancers operating on the device or between the one or more virtual server load balancers on the device and one or more other server load balancers by exchanging information representing load levels.

21. The method of claim 19, wherein instantiating comprises one of instantiating the additional load balancer on the device and instantiating the additional virtual server load balancer on another device.