Abstract: Methods and associated structures for limiting the number of sessions presently active between a server process and client processes distributed over a number of computing devices on a network. The methods and associated structures serve to reduce resource hogging by a particular computing device on a network and thereby reduce opportunities for denial of service attacks on the security of a server node. The server process maintains a session log to record each presently active session between the server process and a computing node identified by a source identifier provided by the client with the establishment and termination of each session. When the number of such active sessions for a particular source identity exceeds the predetermined threshold further requests to establish new sessions associated with that same source identity are rejected by the server process.

Abstract: Methods and associated structure for allocating computing resources among the plurality of cgi-bin processes spawned by a Web server process such that high-priority customers receive priority in allocation of such computing resources for cgi-bin processes its bond on behalf of high-priority customers. The methods and structures of the present invention preferably utilize Hewlett-Packard's process resource monitor (“PRM”) scheduling features to define customer relationship levels as a parameter in the allocation of computing resources. Customer or user ID information associated with a requesting Web client process are associated with the process on behalf of the requesting client. PRM configuration information then mapped such user or customer ID information to a particular priority level for allocation of computing resources.

Abstract: The invention provides a system and method for communicating to a customer device. Briefly described, one embodiment is an access portal configured to relay communications between a customer device and a server device comprising a first access port coupled to a communication system and configured to receive a first communication directed to the access portal from the customer device, the first communication identifying the access portal with an access portal identifier and further requesting content from the server device; a processor configured to map a server device identifier to the received first communication, the server device identifier identifying the server device; and a second access port configured to transmit the first communication with the mapped server device identifier onto a high quality of service communication system such that the first communication is directed to the server device coupled to the high quality of service communication system.

Abstract: Methods and associated structure for allocating computing resources among the plurality of cgi-bin processes spawned by a Web server process such that high-priority customers receive priority in allocation of such computing resources for cgi-bin processes its bond on behalf of high-priority customers. The methods and structures of the present invention preferably utilize Hewlett-Packard's process resource monitor (“PRM”) scheduling features to define customer relationship levels as a parameter in the allocation of computing resources. Customer or user ID information associated with a requesting Web client process are associated with the process on behalf of the requesting client. PRM configuration information then mapped such user or customer ID information to a particular priority level for allocation of computing resources.

Abstract: Methods and associated structures for limiting the number of sessions presently active between a server process and client processes distributed over a number of computing devices on a network. The methods and associated structures serve to reduce resource hogging by a particular computing device on a network and thereby reduce opportunities for denial of service attacks on the security of a server node. The server process maintains a session log to record each presently active session between the server process and a computing node identified by a source identifier provided by the client with the establishment and termination of each session. When the number of such active sessions for a particular source identity exceeds the predetermined threshold further requests to establish new sessions associated with that same source identity are rejected by the server process.