	US 7,543,060 B2
	Service managing apparatus for keeping service quality by automatically allocating servers of light load to heavy task
Takeshi Ishida, Tokyo (Japan); Jinghai Yin, Tokyo (Japan); and Minoru Yamamoto, Tokyo (Japan)
Assigned to Fujitsu Limited, Kawaski (Japan)
Filed on Jul. 03, 2001, as Appl. No. 9/897,100.
Claims priority of application No. 2001-046516 (JP), filed on Feb. 22, 2001.
Prior Publication US 2002/0116479 A1, Aug. 22, 2002
Int. Cl. G06F 15/173 (2006.01)

U.S. Cl. 709—226 [709/223; 709/224]

13 Claims

1. A computer readable storage medium encoded with a computer program causing, when executed by the computer, the computer to perform a service managing method accommodating a plurality of service servers each rendering a service via a network in response to a service request from a client, and distributing the service request to the plurality of service servers, said method comprising:

managing the plurality of service servers by dividing the service servers to define a plurality of groups of service servers depending on quality levels of rendered services, and an intermediate server group of service servers which offer low level service among the service servers at normal time and dynamically shift service servers among the plurality of groups and render a service as a service quality of a group to which the shift is made; and

reducing a load on a service server within any of the plurality of groups by using at least one service server with the lightest load within the intermediate server group as the service server within any of the plurality of groups, when the load on the service server within any of the plurality of groups increases, and a quality level to be rendered by any of the plurality of groups cannot be maintained, and the service server of the intermediate server group returns to a low level when the load on the service server within any of the plurality of groups decreases after the increase of the load, wherein the service request with a high service level requirement is preferentially processed while still processing service requests of a low service level requirement.