Information processing apparatus and job scheduling method
An information processing apparatus calculates a first load value indicating the load of a plurality of first jobs whose used node counts are less than or equal to a threshold among jobs waiting for execution, and a second load value indicating the load of a plurality of second jobs whose used node counts exceed the threshold among the waiting jobs. If load conditions including one where the first load value exceeds the second load value are satisfied, the information processing apparatus combines two or more first jobs among the plurality of first jobs to create a combined job whose used node count exceeds the threshold. The information processing apparatus causes a first node group to execute the remaining first jobs among the plurality of first jobs and causes a second node group to execute the combined job and the plurality of second job.
1. An information processing apparatus comprising:
a memory that holds information on waiting jobs waiting for execution, the waiting jobs each having a used node count specified; and
a processor that performs a process including
calculating a first load value and a second load value, the first load value indicating a first load of a plurality of first jobs whose used node counts are less than or equal to a threshold among the waiting jobs, the second load value indicating a second load of a plurality of second jobs whose used node counts exceed the threshold among the waiting jobs,
combining two or more first jobs among the plurality of first jobs to create a combined job whose used node count exceeds the threshold, in response to a load condition being satisfied, the load condition including a condition that the first load value exceeds the second load value,
causing a first node group to execute remaining first jobs other than the two or more first jobs among the plurality of first jobs, and
causing a second node group to execute the combined job and the plurality of second jobs.
2. The information processing apparatus according to claim 1 , wherein
a scheduled execution time is specified for each of the waiting jobs, and
the first load value is calculated by calculating first products of the used node count and the scheduled execution time with respect to the plurality of first jobs and summing the first products, and the second load value is calculated by calculating second products of the used node count and the scheduled execution time with respect to the plurality of second jobs and summing the second products.
3. The information processing apparatus according to claim 1 , wherein the load condition is that a ratio of the first load value to the second load value exceeds a predetermined ratio greater than one.
4. The information processing apparatus according to claim 1 , wherein the used node count of the combined job is determined based on used node counts of already-executed jobs executed in a past by the second node group.
5. The information processing apparatus according to claim 1 , wherein the two or more first jobs are selected in order of priority from highest from the plurality of first jobs, and the combined job is executed by the second node group preferentially over the plurality of second jobs.
6. A job scheduling method comprising:
calculating, by a processor, a first load value and a second load value, the first load value indicating a first load of a plurality of first jobs whose used node counts are less than or equal to a threshold among waiting jobs waiting for execution, the second load value indicating a second load of a plurality of second jobs whose used node counts exceed the threshold among the waiting jobs;
combining, by the processor, two or more first jobs among the plurality of first jobs to create a combined job whose used node count exceeds the threshold, in response to a load condition being satisfied, the load condition including a condition that the first load value exceeds the second load value;
causing, by the processor, a first node group to execute remaining first jobs other than the two or more first jobs among the plurality of first jobs; and
causing, by the processor, a second node group to execute the combined job and the plurality of second jobs.