Patent · US Active

Failure recovery method in cluster system

US7886181B2 · kind B2 · utility

0Cited by
3References
14Claims
0Family size

Assignee

Inventors

Key dates

Filing dateMar 3, 2008
Grant dateFeb 8, 2011
Priority date
Expiry dateApr 8, 2029

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06F11/2041
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

There is provided a method executed in a cluster system comprising a first computer and at least one second computer that stores the data transmitted from the first computer. The method comprising the steps of: judging whether the failure-occurred processing is restarted by the first computer or the failure-occurred processing is switched over by the second computer based on the system information; receiving the stored data from the second computer and re-executing the processing in which the failure has occurred by the first computer, in the case of which it is judged that the processing in which the failure has occurred is restarted by the first computer; and executing, by the second computer, the processing in which the failure has occurred, in the case of which it is judged that the processing in which the failure has occurred is switched over by the at least one second computer.

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