Year: 2024
Type: conference-paper
Abstract: With the development of integrated circuit technology, the process is gradually approaching the limit, single-core processors have entered the bottleneck period, and multi-core processors have come into being and have received widespread attention. The task scheduling algorithm is one of the key factors in the full utilization of the parallel computing power of multi-core processors. Based on the extensive research on the static task scheduling algorithm of the existing multi-core processor, the replication-based task scheduling algorithm is used as the research object. Aiming at the problems of large critical path estimation error and insufficient solution space search in the existing algorithm, the task replication scheduling algorithm with resource constraints for heterogeneous multicore processors is proposed, which makes the parameter calculation more accurate by supplementing the resource constraints in the parameter calculation process. By supplementing redundant task screening links, unnecessary calculations are reduced during task replication. To verify the performance of the algorithm, the scheduling length of the final layout is selected as the criterion for judging the performance of the algorithm. Through the scheduling comparison experiment of randomly generated task diagram and three practical application task diagrams, the experimental results show that compared with the existing algorithm, the algorithm can improve the parallelism of tasks, shorten the execution time of tasks, and reduce the energy consumption of the system.
Authors: Bin Yuan, Yunya Zhou
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4403717793