Patent · US Active

Fail over method and a computing system having fail over function

US7428210B2 · kind B2 · utility

8Cited by
3References
14Claims
0Family size

Assignee

Inventor

Key dates

Filing dateNov 12, 2004
Grant dateSep 23, 2008
Priority date
Expiry dateAug 29, 2026

Classification

  • Technology area (CPC H)Electricity
  • CPC primaryH04L43/106
  • WIPO fieldDigital communication
  • WIPO sectorElectrical engineering

Abstract

Loads on a plurality of computers are made uniform after a fail over, and resource competition is prevented. Loads of nodes (1 to 4) in a cluster (100) are obtained, taking-over information when a trouble occurs in one of the plurality of nodes (1 to 4) is generated beforehand based on the obtained loads, and the taking-over information is stored in a shared storage system (6) shared by the plurality of nodes. The plurality of nodes (1 to 4) monitor operation states of one another. When an occurrence of a trouble is detected, each of the nodes (1 to 4) in the cluster (100) excluding the troubled node reads the taking-over information from the shared storage system (6), and takes over a processing of the troubled node based on a trouble pattern and the taking-over information.

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