SYSTEM FOR BATCH SCHEDULER USING LOG ANALYSIS BASED ON GRAPH MODEL

摘要

The present invention relates to a batch scheduler system utilizing log analysis based on a graph model. According to the present invention, in a batch scheduler system utilizing log analysis based on a graph model, the workflow meta information is generated based on the workflow using the graph model, and a batch schedule for the generated workflow is executed. Workflow log for generating mining information by analyzing the log information generated by the workflow management unit, the schedule engine unit controlling the workflow management unit based on the workflow meta information, and the execution unit executing the batch schedule based on data mining It includes an analysis unit, a graph-based workflow meta storage unit for storing the workflow meta information and the mining information generated by the workflow log analysis unit, and a schedule execution log storage unit for storing all logs generated by the schedule engine unit do. As described above, according to the present invention, log information is accumulated in the process of executing and managing a batch schedule for a workflow designed based on a graph model, and based on mining information and system resource information obtained by data mining analysis of the accumulated log information to automatically control execution, queuing and prioritization at execution time.