Load balancing

1. A method comprising:

determining, by a network device, that a first number of clients associated with a first access point (AP) is less than a second number of clients associated with a second AP;

setting, by the network device, a target load threshold associated with the first AP, where the target load threshold is based on the second number of clients;

disabling, by the network device, association capabilities, of the first AP, for associating additional clients with the first AP when the first number of clients exceeds the target load threshold; and

applying, by the network device and based on the association capabilities of the first AP being disabled, a limits test to a first coverage area, associated with the first AP, and to a second coverage area, associated with the second AP, where the limits test is applied to determine an area of overlap between the first coverage area and the second coverage area.

2. The method of claim 1 further comprising:

calculating, based on the first number of clients and the second number of clients, an aggregate load associated with the first AP and the second AP; and

redistributing the aggregate load to the first AP and the second AP, where the one or more of the first number of clients are steered to the second AP based on the redistribution of the aggregate load.

3. The method of claim 1 , where applying the limits test further comprises:

interpolating between a limit of the first coverage area and a limit of the second coverage area to determine the area of overlap, where the area of overlap is determined in terms of a smaller of the first coverage area and the second coverage area.

4. A system comprising:

a network device including:

an overlap calculation engine to determine an area of overlap between an access point (AP) and another AP,

a target load calculation engine coupled to the overlap calculation engine, and

an admittance determining engine coupled to the target load calculation engine,

where the target load calculation engine calculates a target load for the AP based on data determined by the overlap calculation engine; and

where the admittance determining engine:

determines whether a number of clients associated with the AP exceeds a threshold number of clients, and

sends, when the number of clients exceeds the threshold number of clients, instructions to the AP to close a safety valve based on the determination, where closing the safety valve causes one or more of clients, seeking to associate with the AP, to be steered to the other AP.

5. The system of claim 4 , where the admittance determining engine:

generates, based on determining whether the number of clients exceeds the threshold number of clients, a value between zero and one that is included in the instructions sent to the AP, where the value comprises zero when the AP is to not admit additional clients, the value comprises one when the AP is to admit the additional clients, and the value is less than one and greater than zero when the AP is to admit some, but not all of the additional clients.

6. The system of claim 4 , where, based on the instructions, clients are steered from the AP to the other AP.

7. The system of claim 4 , where the network device:

receives a message, from a controller associated with an access area that is different than an access area associated with the AP, the message including information about APs in the access area associated with the controller, and

calculates, based on the message, the target load to steer clients to the access area associated with the controller.

8. The system of claim 4 , where a timer limits the duration of the closure of the safety valve to a time determined by the network device.

9. The system of claim 4 , where, based on the calculated target load, the network device transmits a message to a controller, associated with an access area that is different than an access area associated with the AP, to cause the controller to steer clients to the AP.

10. A method comprising:

receiving a report that a first access point (AP), associated with a local access area, is in communication with a second AP, associated with a non-local access area;

identifying a controller associated with the non-local access area;

transmitting a message to the controller;

receiving a message from the controller, the message including a current client load on the second AP;

calculating an overlap between the first AP and the second AP;

calculating a target load for the first AP;

generating, based on determining whether a number of clients associated with the first AP exceeds the target load, a value between zero and one; and

transmitting, to the first AP, instructions including the generated value, where the value comprises zero when the first AP is to not admit additional clients, the value comprises one when the first AP is to admit the additional clients, and the value is less than one and greater than zero when the first AP is to admit some, but not all of the additional clients.

11. The method of claim 10 , where the message is transmitted to the controller based on:

the first AP being in radio communication with the second AP in the non-local access area, controlled by the controller, and

the second AP and the first AP being associated with different access areas.

12. A network device comprising:

a controller to:

determine that a first number of clients associated with a first access point (AP) is less than a second number of clients associated with a second AP,

set, based on the second number of clients, a target load threshold associated with the first AP,

disable, when the first number of clients exceeds the target load threshold, association capabilities of the first AP, for associating additional clients with the first AP,

apply, based on disabling the association capabilities, a limits test to a first coverage area, associated with the first AP, and a second coverage area, associated with the second AP, to determine an area of overlap between the first coverage area and the second coverage area.

13. The network device of claim 12 , where, when disabling the association capabilities, the controller is to:

calculate, based on the target load threshold, a value, and

include the value in instructions transmitted to the first AP, where the value comprises:

a first value to indicate that the first AP is to prevent all of the additional clients from associating with the first AP, or

a second value to indicate that the first AP is to prevent less than all the additional clients from associating with the first AP.