| 摘要 |
PROBLEM TO BE SOLVED: To make a job processing part which has completed the processing of a processing request stop in the case of performing scale-in (the reduction of the number of job processing part) without installing any monitoring device which monitors the state of each job processing part.SOLUTION: When the processing load of a processing group 110 searched from the number of jobs in a job queue 142 becomes less than a threshold, a monitoring part 160 transmits a scale-in instruction to a scale-n queue 146. A job processing part 100 which has received the scale-in instruction first among job processing parts 100 in the processing group 110 stops the acquisition of the job from the job queue 142, and continues the execution of the job during execution at present. Then, when completing all the jobs during execution, the job processing part 100 issues a command to stop itself, and the job processing part 100 disappears as a result. Thus, it is possible to achieve the scale-in of the processing group 110. |
