Jobnet Execution Time Prediction Model for Batch Job Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing batch job management systems face challenges in accurately predicting job execution times, especially with limited past execution data or varying execution times, leading to difficulties in estimating end scheduled times.
Innovation Solution
A job management method and apparatus that generates an execution time prediction model based on historical jobnet execution data, calculates prediction accuracy, and determines the execution end timing of jobnets using this model, displaying predicted information when accuracy meets predetermined criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an execution time prediction model is created based on past execution logs of jobs, then prediction capability is improved, but prediction accuracy deteriorates when there is less past execution time data or large variation in execution time data
Solution Approach 1:
The patent changes the parameter of prediction granularity from individual jobs to jobnets (groups of jobs). By aggregating multiple jobs into jobnets and creating prediction models at the jobnet level, the system obtains more stable prediction data even when individual job execution times vary greatly or have limited historical data. This parameter change from job-level to jobnet-level prediction resolves the contradiction between measurement precision and reliability.
2Loss of information
If prediction is performed for individual jobs, then detailed prediction information is obtained, but prediction accuracy deteriorates due to limited data or large variation in execution times
Solution Approach 1:
The patent merges multiple individual jobs into jobnets (job groups) and performs prediction at the jobnet level. This combining approach aggregates execution data across multiple jobs, providing sufficient statistical basis for accurate prediction while maintaining the ability to track individual job progress within each jobnet. The merging resolves the contradiction by achieving reliable predictions through aggregation while preserving necessary detailed information.
Solution Approach 2:
The patent introduces jobnets as an intermediary layer between individual jobs and the prediction system. Jobnets serve as intermediate units that aggregate job execution data, providing a stable basis for prediction models. This intermediary structure allows the system to overcome the data scarcity and high variability problems of individual jobs while still maintaining awareness of individual job status through the jobnet framework.
Data Source
AI summary
Execution end timing of a jobnet is predicted with stable accuracy. An information processing apparatus executes jobnet execution time prediction model generation processing for generating an execution time prediction model for predicting execution time of a jobnet to be executed on the basis of information associated with execution time of each of previously executed jobnets, a prediction model accuracy determination processing for calculating prediction accuracy for the execution time of each of jobnets by the generated execution time prediction model, and delay determination processing for determining whether to predict execution end timing of a designated jobnet among a jobnet group that is a set of a plurality of jobnets currently being executed or to be subsequently executed on the basis of the execution time prediction model, on the basis of the calculated prediction accuracy for the execution time of each of the jobnets.