摘要 |
PROBLEM TO BE SOLVED: To provide a technique to more surely suppress prolonging of restoration time when a failure occurs in a processing device such as a server.SOLUTION: A management server causes an operation server in which a failure occurs to turn off power (S11), and causes a plurality of standby servers assigned to the operation server to turn on power (S12). Then start completion of a BIOS in the standby servers which are instructed to turn on the power is monitored through a management blade (S13 and S14). After instructing the plurality of standby servers to turn on the power (S22), the management blade 3 monitors start completion of the BIOS in the standby servers (S23 and S24), notifies the management server of a standby server which has completed start of the BIOS most quickly (S25), and instructs the standby server to start an OS (S26). Other standby servers are instructed to turn off the power (S27). |