Failover method in a redundant computer system with storage devices
US7117393B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Apr 30, 2004 |
| Grant date | Oct 3, 2006 |
| Priority date | — |
| Expiry date | Dec 7, 2024 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F11/2084
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A high availability database system, subject to a failure causing data corruption in the active server system, may recover valid data prior to the corruption, and use the protected and recovered data to effect a failover. At the time when a stand-by computer system detects occurrence of a failure, a pair split is done to the data area to protect the data area data of time T0, just before the occurrence of the failure, and another pair split is done to the log area at the time, where the log area is corrupted or a time T3, where the failover is to take effect without corrupting data, to protect the log area containing data at a given time between time T0 and time T3. By using the data and log areas thus protected, the stand-by computer system may take over valid data and log areas to perform a failover.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.