Research on Task Scheduling Algorithm Based on Multi-Time Period Merging

• Published in: 2021 4th World Conference on Mechanical Engineering and Intelligent Manufacturing (WCMEIM) pp. 389 - 394
• Main Authors: Xia, Minghao, Liu, Haibin, Li, Jian, Li, Mingfei
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2021
• Subjects: deadline time; Job shop scheduling; maximum waiting time; Merging; Optimal scheduling; Production; Real-time systems; Robustness; scheduling algorithm; Scheduling algorithms; scheduling time; Task scheduling
• DOI: 10.1109/WCMEIM54377.2021.00085

Task scheduling algorithm is an important component of production task scheduling system. At present, the task scheduling system has some problems related to urgent task scheduling time and complex scheduling logic. Therefore, it is necessary for the task scheduling algorithm to consider the scheduling time and study the logic of the algorithm to ensure the efficient and orderly completion of the scheduling tasks. To solve these problems, this paper proposes a Time Merging Earliest Deadline First (TEDF) algorithm based on the integration of multiple time periods by merging the current time, execution time and task deadline time, which can improve the timeliness of the real-time task scheduling system. In the production task set, the tasks with high static priority are executed first, and in the same priority task queue, the task with the shortest waiting time is executed first as well. The simulation experiment results show that if the task deadline is sufficient, the total waiting time of the tasks processed by the TEDF algorithm is less than that of the ordinary queuing algorithm and the EDF algorithm; if the deadline is limited, the number of timeout tasks and important tasks generated by TEDF algorithm are less than that of ordinary queuing algorithm, which proves that TEDF algorithm can better solve the problems existing in the task scheduling system and has strong robustness.