Server selection method
View Patent ↗A method for selecting ( 64 ) a server ( 30, 40 or 50 ) is based upon an available server capacity indication ( 66 ) for each of the servers. A server may become faulty. Repeated attempted selection ( 63 ) of a faulty server exponentially decreases ( 70 ) the probability of selecting the faulty server. The method performs a ratio of a currently available capacity for each server of a pool of servers to the total available capacity for all the servers in the pool ( 64 ). The server capacities for the pool of servers are dynamically changed ( 68, 70 ).
1. A server selection method for a communication system, the server selection method comprising the steps of:
probabilistically selecting a server by a client;
detecting by the client a fault associated with the server; and
adjusting a probability of selecting the server with the fault including a step of reducing an available capacity indication of the server.
2. The server selection method as claimed in claim 1 , wherein there is further included a step of storing and retrieving by the client the available capacity indication for the server from a database.
3. The server selection method as claimed in claim 1 , wherein there is further included steps of:
providing a plurality of servers; and
providing an available capacity indication for each of the plurality of servers.
4. The server selection method as claimed in claim 1 , wherein there is further included a step of iterating the steps of detecting and adjusting including the step of reducing the available capacity indication for the server with the fault responsive to a plurality of accesses to the server by the client.
5. The server selection method as claimed in claim 1 , wherein the step of reducing the available capacity indication includes a step of minimizing the step of probabilistically selecting the server with the fault.
6. The server selection method as claimed in claim 1 , wherein there is further included a step of requesting by the client a server allocation.
7. The server selection method as claimed in claim 1 , wherein the step of detecting includes a step of determining by the client that a server allocation request has failed.
8. The server selection method as claimed in claim 1 , wherein there is further included steps of:
providing a plurality of servers;
providing an available capacity indication for each of the plurality of servers
selecting one server of the plurality of servers based on the one server with a greatest available capacity indication.
9. The server selection method as claimed in claim 8 , wherein there is further included a step of generating a pseudo-random number for each of the plurality of servers based on a ratio of a currently available capacity indication to a total currently available capacity indication for the plurality of servers.
10. The server selection method as claimed in claim 9 , wherein the step of generating a pseudo-random number for each of the plurality of servers includes a step of generating a pseudo-random number between zero (0) and one (1).
11. The server selection method as claimed in claim 9 , wherein there is further included a step of writing to a database a new currently available capacity indication for each of the plurality of servers.
12. A server selection method for a communication system for selecting one server from a plurality of servers, the server selection method comprising the steps of:
requesting a server by a client from the plurality of servers;
selecting the server based on an available capacity indication for each of the plurality of servers;
detecting by the client a fault associated with the selected server; and
adjusting a probability of selecting the server with the fault including a step of reducing an available capacity indication of the server.
13. The server selection method as claimed in claim 12 , wherein there is further included a step of storing and retrieving by the client the available capacity indication for each of the plurality of servers from a database.
14. The server selection method as claimed in claim 12 , wherein there is further included a step of iterating the steps of selecting, detecting and adjusting including the step of reducing the available capacity indication for the server with the fault responsive to a plurality of accesses by the client.
15. The server selection method as claimed in claim 12 , wherein the step of selecting the server includes a step of minimizing the step of selecting the server with the fault.
16. The server selection method as claimed in claim 12 , wherein the step of selecting the server includes a step of selecting the one server from the plurality of servers based on the server having the greatest available capacity indication.
17. The server selection method as claimed in claim 12 , wherein there is further included a step of generating a pseudo-random number for each of the plurality of servers based on a ratio of a currently available capacity indication to a total currently available capacity indication for the plurality of servers.
18. The server selection method as claimed in claim 17 , wherein the step of generating a pseudo-random number for each of the plurality of servers includes a step of generating a pseudo-random number between zero (0) and one (1).
19. The server selection method as claimed in claim 12 , wherein there is further included a step of writing to a database a new currently available capacity indication for each of the plurality of servers.
20. A server selection method for a communication system for selecting one server from a plurality of servers, the server selection method comprising the steps of:
selecting the one server based on a currently available capacity indication for each of the plurality of servers;
detecting by the client a fault associated with the selected server;
adjusting a probability of selecting the server with the fault including a step of reducing the currently available capacity indication of the server; and
iterating by the client the steps of detecting and adjusting including the step of reducing the currently available capacity indication for the server with the fault responsive to lower a probability of selecting the one server having the fault.