Parallel processing apparatus and method of estimating power consumption of jobs
A processor extracts a first file name from a file path indicating a storage location of a first file used when launching a first job. The processor calculates, for each of a plurality of second jobs that have been executed, a similarity based on a comparison between partial character strings included in the first file name and partial character strings included in a second file name corresponding to a second file that was used when launching the second job. The processor acquires, from history information indicating an actual power consumption of each of the plurality of second jobs and in accordance with the calculated similarity, the actual power consumption of at least one second job and estimates power consumption of the first job based on the acquired actual power consumption.
1. A non-transitory computer-readable storage medium storing a computer program, the computer program that causes a computer to perform a procedure comprising:
extracting a first file name, which is part of a file path indicating a storage location of a first file used when launching a first job, from the file path;
calculating, for each of a plurality of second jobs that have been executed by using respective second files, similarity based on a comparison between partial character strings included in the first file name and partial character strings included in a second file name of a corresponding one of the second files; and
acquiring, from history information indicating an actual power consumption of each second job in the plurality of second jobs and in accordance with the calculated similarity, the actual power consumption of at least one second job and estimating power consumption of the first job based on the acquired actual power consumption.
2. The non-transitory computer-readable storage medium according to claim 1 , wherein the first file name includes a first directory name indicating a directory on a level immediately above the first file and the second name includes a second directory name indicating a directory on a level immediately above the corresponding one of the second files.
3. The non-transitory computer-readable storage medium according to claim 1 , wherein the calculating of the similarity includes comparing a set of n consecutive characters included in the first file name with a set of n consecutive characters included in the second file name, where n is an integer of 2 or higher.
4. A parallel processing apparatus comprising:
a memory that stores history information indicating, for each of a plurality of second jobs that have been executed by using respective second files, a second file name of a corresponding one of the second files and actual power consumption of the second job corresponding to the second file name; and
a processor that extracts a first file name, which is part of a file path indicating a storage location of a first file used when launching a first job, from the file path, calculates, for each of the plurality of second jobs, similarity based on a comparison between partial character strings included in the first file name and partial character strings included in the second file name, acquires the actual power consumption of at least one second job from the history information in accordance with the calculated similarity, and estimates power consumption of the first job based on the acquired actual power consumption.
5. A method of estimating power consumption of a job, the method comprising:
extracting, by a processor, a first file name, which is part of a file path indicating a storage location of a first file used when launching a first job, from the file path;
calculating, by the processor and for each of a plurality of second jobs that have been executed by using respective second files, similarity based on a comparison between partial character strings included in the first file name and partial character strings included in a second file name of a corresponding one of the second files; and
acquiring, by the processor, in accordance with the calculated similarity, and from history information indicating an actual power consumption of each second job in the plurality of second jobs, the actual power consumption of at least one second job and estimating, by the processor, power consumption of the first job based on the acquired actual power consumption.