Algorithm for Improving Processor Utilization in Multi-core Processor Environment by Python Language

• Published in: IEEE Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC ...) (Online) Vol. 4; pp. 775 - 779
• Main Authors: He, Feng, Hu, Xuran, Liu, Shuren, Li, Tao, Zhu, Kaifeng, Bao, Xin, Jiang, Chao
• Format: Conference Proceeding
• Language: English
• Published: IEEE 18.06.2021
• Subjects: computing-intensive tasks; Concurrent computing; GIL; Information management; Machine learning algorithms; multi-process; multi-thread; Multicore processing; Natural language processing; Process control; Python; signal mechanism; Writing; zombie process
• ISSN: 2693-2776
• DOI: 10.1109/IMCEC51613.2021.9481962

Python is a very popular programming language at present, and it has a wide range of applications in big data applications and analysis, artificial intelligence and other fields. Especially in recent years, the application of Python in the field of artificial intelligence has involved natural language processing, computer vision, and machine learning. When writing computation-intensive tasks, such as image recognition programs based on deep learning, this paper briefly analyzes a concise algorithm that uses a combination of multi-process and signal mechanisms, and uses queues to share data between processes, thereby making full use of CPU and GPU performance. The algorithm can clean up zombie processes in time, make better use of system resources, and meet the requirements of multi-user concurrency. It also explains why this strategy is used and how to use it. Experiments show that the algorithm has good effect.