Controlling the operation of server computers
View Patent ↗A plurality of server computers in a network that includes load balancing processes to enhance performance employs a non-performance related variable, such as power consumption, and modifies the load balancing processes in response to the reading of the non-performance variable. Such variables do not affect the response and performance as perceived by a browsing client but do provide other advantages with in the environment as a whole. The non-performance related variable is employed such that one or more of said server computers are de-activated to reduce power consumption, and the load balancing processes balance load across the remaining active server computers.
1. A load balancing system for balancing load based on performance based factors and non-performance based factors, the load balancing system comprising:
a first local network, in a first geographic location, including a first plurality servers;
a second local network, in a second geographic location, including a second plurality of servers;
a traffic manager communicatively coupled to the first local network and the second local network, the traffic manager comprising:
a non-transitory computer-readable medium having instructions stored thereon that when executed cause the traffic manager to:
receive a request for data from a requesting client device, wherein the request for data can be served from the first plurality of servers or the second plurality of servers;
identify candidate servers to service the request for data from the first and second plurality of servers based on a load balancing performance variable, wherein the load balancing performance variable is one that affects a response and performance as perceived by the requesting client device; and
identify a first subset of the candidate servers based on a first non-performance variable;
identify a second subset of the first subset of the candidate servers based on a second non-performance variable that is different than the first non-performance variable; and
select one of the second subset of the first subset of the candidate servers to provide a service based on the load balancing performance variable, the non-performance variable, and the second non-performance variable.
2. The system of claim 1 , wherein the first non-performance variable or the second non-performance variable is time of day and location, whereby the traffic manager selects the first subset of candidate servers or the second subset of candidate servers based on a scheduled daily cycle such that the subset of servers are those that are preferred for the time of day that the request for data is received.
3. The system of claim 1 , wherein the first non-performance variable or the second non-performance variable is an economic factor, whereby the traffic manager selects the first subset of candidate servers or the second subset of candidate servers that achieve a lower economic cost than servers outside the subset of candidate servers.
4. The system of claim 3 , wherein the load balancing performance variable is capacity.
5. The system of claim 1 , wherein the first non-performance variable or the second non-performance variable is a duty cycle of the servers, whereby the traffic manager selects the first subset of candidate servers or the second subset of candidate servers that minimizes a number of active servers.
6. A method for operating a load balancing system to optimize performance an economic performance, comprising:
receiving a request for data from a requesting client device, wherein the request for data can be served from a system of servers, wherein the system of servers includes a first plurality of servers located at a first geographic location, and a second plurality of servers located at a second geographic location;
determining by a traffic manager a multiple servers from the system of servers that has capacity to service the request;
identifying from among the multiple servers having capacity to service the request based on a non-performance factor to yield identified server(s); and
select from among the identified servers based on a reconsideration of the capacity to yield selected server(s) to provide a service based on the capacity and the non-performance variable.
7. The method of claim 6 , further comprising:
when the selected sever(s) number two or more, selecting from among the selected servers based on a second non-performance variable;
selecting one of two or more selected servers to provide the service based on the capacity to service the request, the non-performance variable, and the second non-performance variable.
8. The method of claim 6 , wherein the non-performance variable is time of day and location, whereby the traffic manager selects the selected sever(s) based on a scheduled daily cycle such that the selected sever(s) are those that are preferred for the time of day that the request for data is received.
9. The method of claim 6 , wherein the non-performance variable is an economic factor, whereby the traffic manager selects the selected sever(s) that achieve a lower economic cost than servers that are not the selected sever(s).
10. The method of claim 6 , wherein the non-performance variable is a duty cycle of the servers, whereby the traffic manager selects the selected sever(s) that minimizes the number of active servers.