Patent · US Active

Task allocation method and task allocation apparatus

US7930339B2 · kind B2 · utility

40Cited by
5References
12Claims
0Family size

Assignee

Inventors

Key dates

Filing dateNov 14, 2006
Grant dateApr 19, 2011
Priority date
Expiry dateNov 28, 2028

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06F2209/483
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A distributed processing system includes plural nodes each provided with a processor and communicably connected to each other. A task information obtaining unit obtains antecedent dependency between tasks. A node information obtaining unit obtains a time constraint imposed on the tasks. A start time computing unit computes an absolute earliest start time at which a task can be started and an absolute latest start time allowed in order to complete the task within the time constraint, by referring to the time constraint. A node selector selects a destination node to which a target task not yet allocated to a node should be allocated, by referring to the absolute earliest start time and the absolute latest start time. A task placement unit causes the task to be executed in the selected node.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.