METHODS AND APPARATUS FOR PERFORMING DYNAMIC LOAD BALANCING OF PROCESSING RESOURCES

• US 20140101226A1
• Filed: 10/08/2012
• Published: 04/10/2014
• Est. Priority Date: 10/08/2012
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola Mobility LLC (Lenovo Group Ltd.)
• Inventors: Khandekar, Milind M., Jeyakeerthi, Jeyappragash J.
• Granted Patent: US 9,071,609 B2
• Time in Patent Office: Days
• US Class Current: 709/203
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: G06F 9/5055 considering software capabilities, i.e. software resources associated or available to the machine
• 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 67/1029 using data related to the state of servers by a load balancer
• 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
• CPC Class Codes: H04L 67/56 Provisioning of proxy services store-and-forward switching systems in data switching networks H04L12/54

Abstract

A load balancer coupled between a network and a pool of processing resources utilizes a method for performing automatic, dynamic load balancing with respect to allocating the resources to handle processing requests received over the network. According to the method, the load balancer electronically receives registration requests from the processing resources and registers each resource from which a registration request was received. After the resources have been registered but before they are allocated to handle processing requests, the load balancer receives, from each registered resource, information relating to utilization of the resource. The utilization information may include operational metrics for the resource. Some time thereafter, the load balancer receives a request over the network to perform a server-related processing task. Responsive to the request, the load balancer allocates at least one of the registered resources to perform the requested processing task based at least on the previously received utilization information.

Claims

1. A method for a load balancer to manage allocation of a plurality of processing resources to perform server-related processing tasks requested by client applications over a network, the method comprising:

electronically receiving registration requests from at least some of the plurality of processing resources;

registering each processing resource from which a registration request was received to produce registered processing resources;

receiving, from each registered processing resource, information relating to utilization of the registered processing resource to collectively produce utilization information;

receiving a request to perform a server-related processing task from a client application over the network; and

responsive to the request to perform the server-related processing task, allocating at least one of the registered processing resources to perform the server-related processing task based on the utilization information.

2. The method of claim 1, further comprising:

electronically communicating address information for the load balancer to at least one address resolution server on the network, the address information enabling the plurality of processing resources to determine an electronic address of the load balancer through communication with the at least one address resolution server;

wherein electronically receiving registration requests from at least some of the plurality of processing resources occurs subsequent to communicating the address information to the at least one address resolution server.

3. The method of claim 1, wherein receiving information relating to utilization of the registered processing resource comprises:

remotely executing a script at the registered processing resource, the script enabling the load balancer to retrieve the utilization information from the load balancer.

4. The method of claim 3, further comprising:

receiving information regarding the script from the registered processing resource during registration with the load balancer.

5. The method of claim 1, wherein receiving information relating to utilization of the registered processing resource comprises:

remotely executing a script at the registered processing resource that enables the load balancer to retrieve a data structure from the registered processing resource, wherein the data structure includes the information relating to utilization of the registered processing resource.

6. The method of claim 1, wherein the information relating to utilization of the registered processing resource includes operational metrics for the registered processing resource.

7. The method of claim 6, wherein the operational metrics include at least one of memory used, network input and output ports, central processing unit usage, swap memory usage, operating system cache, and quantity of processor operations executed.

8. The method of claim 1, further comprising:

remotely executing a notification script at a particular registered processing resource, the notification script notifying the particular registered processing resource of an event related to the load balancer.

9. The method of claim 1, further comprising:

electronically receiving a deregistration request from a registered processing resource; and

deregistering the registered processing resource from which the deregistration request was received.

10. The method of claim 1, wherein the plurality of processing resources includes at least one of:

servers, one or more cloud computing environments, and one or more additional load balancers.

11. The method of claim 1, wherein receiving, from each registered processing resource, information relating to utilization of the registered processing resource comprises:

receiving, from a cloud computing management system application, information relating to utilization of server instances in an associated cloud computing environment when the plurality of processing resources includes the cloud computing environment.

12. The method of claim 11, wherein the cloud computing management system application is an orchestration layer.

13. The method of claim 1, wherein the load balancer is implemented as at least one of a hardware device and a software process within a host computing system coupled to the network.

23. The method of claim 1, further comprising:

electronically communicating a deregistration request to the load balancer after the processing resource has been registered with the load balancer.

14. A computing system for implementing a load balancer, the computing system comprising:

one or more interfaces for interfacing the computing system to one or more networks and to a plurality of processing resources; and

a controller coupled to the one or more interfaces, the controller being configured to;

electronically receive registration requests from at least some of the plurality of processing resources;

register each processing resource from which a registration request was received to produce registered processing resources;

receive, from each registered processing resource, information relating to utilization of the registered processing resource to collectively produce utilization information;

receive a request to perform a server-related processing task from a client application over a network; and

responsive to the request to perform the server-related processing task, allocate at least one of the registered processing resources to perform the server-related processing task based on the utilization information.

15. The computing system of claim 14, wherein the one or more networks include at least one address resolution server and wherein the controller is further configured to:

electronically communicate address information for the load balancer to the at least one address resolution server, the address information enabling the plurality of processing resources to determine an electronic address of the load balancer through communication with the at least one address resolution server; and

electronically receive the registration requests from at least some of the plurality of processing resources subsequent to communication of the address information to the at least one address resolution server.

16. The computing system of claim 15, wherein the at least one address resolution server comprises at least one domain name system server.

17. The computing system of claim 14, wherein the plurality of processing resources includes a cloud computing environment, wherein allocation of server instances in the cloud computing environment is controlled by a cloud computing management system application, and wherein the controller is configured to receive information relating to utilization of the server instances in the cloud computing environment from the cloud computing management system application.

18. A computer-readable storage medium including instructions that, when executed by a computing system, cause the computing system to:

electronically receive registration requests from at least some of a plurality of processing resources;

register each processing resource from which a registration request was received to produce registered processing resources;

receive, from each registered processing resource, information relating to utilization of the registered processing resource to collectively produce utilization information;

receive a request to perform a server-related processing task from a client application over a network to which the computing system is communicatively coupled; and

responsive to the request to perform the server-related processing task, allocate at least one of the registered processing resources to perform the server-related processing task based on the utilization information.

19. A method for a processing resource to assist a load balancer in performing dynamic load balancing, the processing resource being one of a plurality of processing resources allocatable by the load balancer and capable of performing server-related processing tasks as requested by remotely-executing client applications, the method comprising:

generating information relating to utilization of the processing resource to produce utilization information;

performing a registration sequence with the load balancer;

subsequent to the registration sequence, receiving a request for the utilization information from the load balancer; and

communicating the utilization information to the load balancer responsive to the request.

20. The method of claim 19, further comprising:

prior to performing the registration sequence;

accessing an address resolution server on a network to which the load balancer is communicatively coupled;

determining an address of the load balancer from the address resolution server; and

communicating a registration request to the address of the load balancer as determined from the address resolution server.

21. The method of claim 19, further comprising:

communicating information relating to a script to the load balancer during the registration sequence, wherein the script resides in a processing space of the processing resource and, when remotely executed by the load balancer, notifies the processing resource of an event related to the load balancer; and

executing the script responsive to instructions from the load balancer.

22. The method of claim 19, further comprising:

communicating information relating to a script to the load balancer during the registration sequence, wherein the script resides in a processing space of the processing resource and, when remotely executed by the load balancer, causes the processing resource to communicate the utilization information to the load balancer; and

executing the script responsive to an instruction from the load balancer, wherein the instruction corresponds to the request for the utilization information.