Heterogeneous platform load balancing scheduling method and system based on improved ant colony

• Main Authors: YUE CHUNSHENG, ZHOU NAN, MA JINQUAN, LI YUDONG, HU ZEMING, ZHANG KUN
• Format: Patent
• Language: Chinese; English
• Published: 06.02.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention relates to the technical field of heterogeneous platform task scheduling, in particular to a heterogeneous platform load balancing scheduling method and system based on an improved ant colony, task types are divided into a communication intensive type and a calculation intensive type, and a heterogeneous platform task priority scheduling list is generated by utilizing shunt sorting; taking a to-be-processed task set of the heterogeneous platform and the state of each processor of the current heterogeneous platform as a state space of Q learning, taking a process of allocating tasks to the processors as an action space, and obtaining an initial task hardware resource allocation strategy by utilizing a Q learning algorithm and according to the execution time required by mapping the tasks to the action space in the Q learning; and taking an initial task hardware resource allocation strategy as an ant colony initial pheromone, describing mapping between a task and a processor in a task scheduling sc