Systems and methods for load balancing based on user selected metrics

• US 8,291,108 B2
• Filed: 03/12/2007
• Issued: 10/16/2012
• Est. Priority Date: 03/12/2007
• Status: Active Grant

• Current Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Original Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Inventors: Raja, Murali, Muthiah, Manikam, Kunal, Ranveer, Raghav, Somanahalli Narayana
• Primary Examiner(s): Avellino, Joseph
• Assistant Examiner(s): Afolabi, Mark O
• Application Number:
  US11/685,185
• Publication Number:
  US 20080225710A1
• Time in Patent Office: 2,045 Days
• Field of Search: 709/220, 709/201, 709/219, 709/217, 709/218, 709/249, 709/238, 709/232, 709/224, 709/226, 709/203, 709/245, 707/501, 707/505, 707/506, 707/530, 707/513, 715/215, 715/735, 715/736, 715/740, 370/230, 370/231, 370/232, 370/352, 370/389, 370/229, 370/254, 370/360, 705/5, 705/6, 705/1, 705/28, 705/26, 705/27, 705/56, 705/35, 705/37, 705/44, 705/10, 705/14, 705/30, 705/7
• US Class Current: 709/238
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: H04L 45/00 Routing or path finding of packets in data switching networks routing or path finding in wireless networks H04W40/00
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/125 by balancing the load, e.g. traffic engineering
• CPC Class Codes: H04L 47/193 at the transport layer, e.g. TCP related
• CPC Class Codes: H04L 47/20 Traffic policing
• CPC Class Codes: H04L 47/29 using a combination of thresholds
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/101 based on network conditions
• CPC Class Codes: H04L 67/1012 based on compliance of requirements or conditions with available server resources
• CPC Class Codes: H04L 67/1017 based on a round robin mechanism
• CPC Class Codes: H04L 67/1029 using data related to the state of servers by a load balancer

Abstract

The present invention provides improvements to load balancing by providing a load balancing solution that allows a user to select metrics, weights and thresholds from metrics collected or determined by a load balancing appliance as well as metrics obtained from another device, such as a server running the service. The systems and methods described providing a load balancing technique in which load balancing decisions can be made based on metrics known by another device monitoring such metrics. The load balancing appliance uses a network management protocol and communication model, such as a Simple Network Management Protocol (SNMP), to identify and collect metrics from a server, service or other device. These metrics are available on the load balancing appliance with appliance determined metrics. Via a configuration interface of the appliance, a user can select one or more appliance determined metrics and/or the metrics obtained via the network management protocol to determine the load of a service.

Claims

1. A method of performing load balancing of a service based on a user selected metric, the method comprising:

identifying, via an appliance in communication via a network with a plurality of heterogeneous load balancing devices for which the appliance provides global server load balancing, a plurality of different metrics provided by the plurality of heterogeneous load balancing devices for a plurality of services to be load balanced by the appliance, the plurality of heterogeneous devices comprising devices of different types;

receiving, by the appliance, information identifying selection by a user of a first metric of a first heterogeneous load balancing device of the plurality of heterogeneous load balancing devices from a plurality of metrics provided by the first heterogeneous load balancing device

receiving, by the appliance, a first threshold designated by the user for the first user selected metric;

determining, by the appliance from monitoring the first user selected metric collected from the first heterogeneous load balancing device, a load for the plurality of services based on the first user selected metric and the first user designated threshold;

receiving, by the appliance, a request of a client for a service;

determining, by the appliance, that the first user selected metric for a first service of the plurality of services has reached the user identified threshold; and

selecting, by the appliance, a second service from the plurality of services to receive the request.

2. The method of claim 1, comprising monitoring, by one or more load monitors of the appliance, a plurality of virtual servers managing the plurality of services.

3. The method of claim 2, comprising determining, by a load monitor, that the first user selected metric for a virtual server exceeds the predetermined threshold.

4. The method of claim 3, comprising excluding, by the appliance, the virtual server from one of the determination of the load or selection for handling client requests.

5. The method of claim 1, comprising receiving, by the appliance, information identifying selection by the user of a second metric from the plurality of metrics provided by a second heterogeneous load balancing device of the plurality of heterogeneous load balancing devices.

6. The method of claim 5, comprising determining, by the appliance, the load for the plurality of services based on the first user selected metric and the second user selected metric.

7. The method of claim 6, comprising receiving, by the appliance, information identifying a second threshold designated by the user for the second metric.

8. The method of claim 1, comprising receiving, by the appliance, information identifying a selection by the user of a first weighting factor for the first user selected metric.

9. The method of claim 1, comprising establishing, by the appliance, an appliance determined weighting factor for the first user selected metric.

10. The method of claim 1, comprising receiving, by the appliance, a request of a client for a service, and selecting, by the appliance, a first service from the plurality of services to receive the request based on the determined load.

11. The method of claim 1, comprising determining, by the appliance, that the first user selected metric for a service is within the user selected threshold.

12. The method of claim 1, comprising receiving, by the appliance, a request of a client for a service, determining that the first user selected metric for a first service of the plurality of services has reached the user identified threshold, and redirecting the request of the client to one of the following:

a second appliance, a server, and a second set of one or more services.

13. The method of claim 1, comprising receiving, by the appliance, a request of a client for a service, determining that the first user selected metric for a virtual server exceeds the user identified threshold, and providing the client a redirect uniform resource locator to one of the following:

a second appliance, a server, and a service.

14. The method of claim 1, comprising obtaining, by the appliance, a value of one of the first user selected metric or a second metric via a Simple Network Management Protocol (SNMP) query to a device.

15. The method of claim 1, comprising determining, by the appliance, a value of one of the first user selected metric based or a second metric based on an average of collected values for the metric.

16. The method of claim 1, comprising indicating, by the first user selected metric, one of a performance or an operational characteristic of a service.

17. The method of claim 1, wherein the plurality of metrics comprises one of the following:

a number of connections to a service, a number of packets sent to or transmitted by the service, a response time of the service to respond to a request, and use of network bandwidth by the service.

18. The method of claim 1, comprising receiving, by the appliance, information from the user specifying to the appliance to collect a user-defined metric.

19. The method of claim 18, comprising receiving, by the appliance, information from the user identifying an object identifier for the user-defined metric to query via a Simple Network Management Protocol (SNMP) request to a server.

20. A method of performing load balancing of a service based on a user selected metric, the method comprising:

receiving, by the appliance in communication via a network with a plurality of heterogeneous load balancing devices providing different metrics identified via the appliance and for which the appliance provides global server load balancing, a first metric selected by a user from a plurality of metrics provided by a heterogeneous load balancing device of the plurality of heterogeneous load balancing devices for a service to be load balanced by the appliance and a first threshold for the first metric designated by the user, the plurality of heterogeneous load balancing devices comprising devices of different types;

establishing, by the appliance, a second metric of a second plurality of metrics provided by the appliance different from the first metric of the plurality of metrics provided by the heterogeneous load balancing device and a second threshold for the second metric for load balancing the service;

collecting, by the appliance, data for the first user selected metric from the heterogeneous load balancing device and data for the appliance established second metric provided by the appliance;

determining, by the appliance, a load for the service based on the first user selected metric collected from the heterogeneous load balancing device and the first user designated threshold and the appliance established second metric collected via the appliance and the second threshold;

receiving, by the appliance, a request of a client for a service;

determining, by the appliance, that one of the user selected metric or the second metric for a first service of the plurality of services has reached the user identified threshold; and

selecting, by the appliance, a second service from the plurality of services to receive the request.

21. The method of claim 20, receiving, by the appliance, a request of a client for a service, and selecting, by the appliance, a first service from the plurality of services to receive the request based on the determined load.

22. The method of claim 20, comprising determining, by the appliance, that the second metric is within an appliance established threshold.

23. The method of claim 20, comprising determining, by the appliance, that the first user selected metric for the service is within the user selected threshold.

24. The method of claim 20, comprising determining, by the appliance, the first user selected metric for the service exceeds the user selected threshold and excluding the service from the determination of the load.

25. The method of claim 20, further comprises receiving, by the appliance, a weighting factor designated by the user for one of the first user selected metric or the second metric established by the appliance.

26. The method of claim 25, comprising determining, by the appliance, the load for the service based on the user designated weighting factor and the first user designated threshold.

27. The method of claim 20, comprising receiving, by the appliance, a request of a client for a service, determining that one of the user selected metric or the second metric for a first service of the plurality of services has reached the user identified threshold, and redirecting the client to one of the following:

a second appliance, a server, and a second set of one or more services.

28. The method of claim 20, comprising receiving, by the appliance, a request of a client for a service, determining that the first user selected metric for virtual server exceeds the user identified threshold, and providing the client a redirect uniform resource locator to one of the following:

a second appliance, a server, and a service.

29. The method of claim 20, further comprises obtaining, by the appliance, a value of one of the first user selected metric or the second metric via a Simple Network Management Protocol (SNMP) query to a server.

30. The method of claim 20, further comprises obtaining, by the appliance, a value of one of the second metric by monitoring requests and responses to the service.

31. The method of claim 20, comprising indicating, by one of the first user selected metric or the second metric, one of a performance or an operational characteristic of a service.

32. The method of claim 20, wherein the plurality of metrics comprises one of the following:

a number of connections to a service, a number of packets sent to or transmitted by the service, a response time of the service to respond to a request, and use of network bandwidth by the service.

33. An appliance for performing load balancing of a service based on a user selected metric and user designated threshold, the appliance comprising:

a configuration interface for receiving information identifying a first metric of a first heterogeneous load balancing device of a plurality of heterogeneous load balancing devices selected by a user from a plurality of different metrics identified as provided from the plurality of heterogeneous load balancing devices for load balancing a plurality of services, and information identifying a first threshold designated by the user for the first metric;

a virtual server for directing requests from clients to the plurality of services;

a load monitor collecting data for the first user selected metric from the first heterogeneous load balancing device;

wherein the load monitor determines a load for the service based on the first user selected metric and the first user designated threshold and the virtual server responsive to the load monitor directs a client request to a service of the plurality of services; and

wherein the appliance receives a request of a client for a service, the load monitor determines that the first user selected metric for a first service of the plurality of services has reached the user identified threshold, and selecting by the virtual server a second service from the plurality of services to receive the request server a second service from the plurality of services to receive the request.

34. The appliance of claim 33, wherein the load monitor determines that the first user selected metric for a virtual server exceeds the predetermined threshold.

35. The appliance of claim 34, wherein the appliance excludes the virtual server from one of the determination of the load or selection for handling client requests.

36. The appliance of claim 33, wherein the configuration interface receives information identifying selection by the user of a second metric from the plurality of metrics.

37. The appliance of claim 36, wherein the load monitor determines the load for the plurality of services based on the first user selected metric and the second user selected metric.

38. The appliance of claim 37, wherein the configuration interface receives information identifying a second threshold designated by the user for the second metric.

39. The appliance of claim 33, wherein the configuration interface receives information identifying a selection by the user of a first weighting factor for the first user selected metric.

40. The appliance of claim 33, wherein the load monitor establishes a weighting factor for the first user selected metric.

41. The appliance of claim 33, wherein the appliance receives a request of a client for a service, and the virtual server selects the service from the plurality of services to receive the request based on the determined load.

42. The appliance of claim 33, wherein the load monitor determines that the first user selected metric for a service is within the user selected threshold.

43. The appliance of claim 33, wherein the appliance receives a request of a client for a service, the load monitor determines that the first user selected metric for a service of the plurality of services has reached the user identified threshold, and the appliance redirects the request of the client to one of the following:

a second appliance, a server, and a second set of one or more services.

44. The appliance of claim 33, wherein the appliance receives a request of a client for a service, the load monitor determines that the first user selected metric for the virtual server exceeds the user identified threshold, and the appliance, responsive to the load monitor, provides the client a redirect uniform resource locator to one of the following:

a second appliance, a server, and a service.

45. The appliance of claim 33, wherein the load monitor obtains a value of one of the first user selected metric or a second metric via a Simple Network Management Protocol (SNMP) query to a device.

46. The appliance of claim 33, wherein the load monitor obtains a value of one of the first user selected metric based or a second metric based on an average of collected values for the metric.

47. The appliance of claim 33, wherein the first user selected metric indicates one of a performance or an operational characteristic of a service.

48. The appliance of claim 33, wherein the plurality of metrics comprises one of the following:

a number of connections to a service, a number of packets sent to or transmitted by the service, a response time of the service to respond to a request, and use of network bandwidth by the service.

49. The appliance of claim 33, wherein the configuration interface receives information from the user specifying to the appliance to collect a user-defined metric.

50. The appliance of claim 49, wherein the configuration interface receives information from the user identifying an object identifier for the user-defined metric to query via a Simple Network Management Protocol (SNMP) request to a server.