System for balancing loads among network servers

  • US 6,128,279 A
  • Filed: 10/01/1998
  • Issued: 10/03/2000
  • Est. Priority Date: 10/06/1997
  • Status: Expired due to Term
First Claim
Patent Images

1. A method of distributing requests among a plurality of network servers, the method comprising the steps of:

  • receiving a request from a remote source at a first one of the network servers;

    executing a determining step in the first server, the determining step for determining whether to process the request in the first network server;

    processing the request in the first network server in a case that the determining step determines that the request should be processed in the first network server; and

    routing the request to another network server in a case that the determining step determines that the request should not be processed in the first network server;

    wherein the determining step comprises the steps of;

    determining a load currently being processed by the first network server; and

    receiving information in the first network server from each of the other network servers, the information from each of the other network servers comprising information concerning a load currently being processed in each network server;

    wherein the determining step determines that the first network server should process the request in a case that (i) the load currently being processed in the first network server is below a first predetermined level, or (ii) the load currently being processed in the first network server is above the first predetermined level and is above loads currently being processed by either of the other network servers by less than a second predetermined level; and

    wherein the determining step determines that the first network server should not process the request in a case that the load currently being processed in the first network server is above the first predetermined level and a load currently being processed in at least one of the other network servers is below the level of the first network server by at least the second predetermined level.

View all claims
    ×
    ×

    Thank you for your feedback

    ×
    ×