Information processing apparatus and job scheduling method

A job scheduling method, comprising:

• performing by at least one computer:
• storing in at least one memory, a first execution time and a second execution time longer than the first execution time, the first and second execution times being different execution durations based upon different time periods expected to execute a first job included in a first job group;
• determining in which execution time slot, from among execution time slots, to schedule an execution start time of the first job to start execution of the first job;
• selecting, as a predicted execution duration of the first job, one of the first execution time and the second execution time based on the execution time slot determined to schedule the execution start time of the first job;
• schedule the first job group based on the predicted execution duration,
• set job flow information including information indicating a maximal flow of jobs to be completed within a period of time as a currently set critical path, and a predicted execution duration to execute a flow of lobs being maximal for the maximal flow among flows of scheduled jobs;
• execute lobs included in the first job group while giving a job present on the critical path a higher priority than a priority given to a job not present on the critical path;
• compare, after execution of the first job is ended, the predicted execution duration of the first job with an execution time taken for the execution of the first job;
• and
• evaluate, based on a result of the comparison, whether the currently set critical path is still maximal to set a new critical path.

A non-transitory computer-readable recording medium having stored therein a program for causing a computer to execute a process, the process comprising:

• storing in at least one memory, a first execution time and a second execution time longer than the first execution time, the first and second execution times being different execution durations based upon different time periods expected to execute a first job included in a first job group;
• determining in which execution time slot, from among execution time slots, to schedule an execution start time of the first job to start execution of the first job;
• selecting, as a predicted execution duration of the first job, one of the first execution time and the second execution time based on the execution time slot determined to schedule the execution start time of the first job;
• scheduling the first job group based on the predicted execution duration,
• setting job flow information including information indicating a maximal flow of lobs to be completed within a period of time as a currently set critical path, and a predicted execution duration to execute a flow of lobs being maximal for the maximal flow among flows of scheduled jobs;
• executing lobs included in the first job group while giving a job present on the critical path a higher priority than a priority given to a job not present on the critical path;
• comparing, after execution of the first job is ended, the predicted execution duration of the first job with an execution time taken for the execution of the first job;
• and
• evaluating, based on a result of the comparing, whether the currently set critical path is still maximal to set a new critical path.