International Business Machines Corporation

Steven Armand Jarvis

Liliana Orozco

Jeffrey Mark Achtermann

Brian George Vassberg

Douglas Ledden

A mechanism for providing efficient use of peer cache space in large scale file distributions. A peer computer receives a download plan for a requested file. The download plan specifies a number of peer servers where the file is stored, a minimum number of peer servers required to store the file, and a saturation level for the file in the zone. When the peer computer downloads the file from one or more peer servers listed in the download plan, the peer computer determines whether the file is required to be cached at the peer computer by comparing the minimum number of peer servers in the plan required to store the file against the number of peer servers listed in the plan. If the minimum number required to store the file is greater the number of peer servers listed in the plan, the peer computer caches the file.

Efficient use of peer cache space in large scale file distributions

A system, computer-implementable method, and computer-readable medium for processing a collection of work items in a data processing system. According to a preferred embodiment of the present invention, an agent monitor assembles a work items list that includes a collection of work items, wherein at least one work item among the collection of work items is a work item that is unassigned to at least one agent among a collection of agents. The agent monitor assigns the at least one work item to the at least one agent. In response to determining that the at least one agent has entered into a hang state, the agent monitor releases the at least one work item from the at least one agent.

Growing community of inventors

Austin, TX, United States of America

Steven Armand Jarvis