A Genetic Programming-Based Approach for Automated Dispatching Rules Design in Dynamic EDA Scientific Workflows

Created by W.Langdon from gp-bibliography.bib Revision:1.8620

@InProceedings{DBLP:conf/cec/SunM25,

• author = "Yu Sun and Ying Meng",
• title = "A Genetic Programming-Based Approach for Automated Dispatching Rules Design in Dynamic {EDA} Scientific Workflows",
• booktitle = "2025 IEEE Congress on Evolutionary Computation (CEC)",
• year = "2025",
• editor = "Yaochu Jin and Thomas Baeck",
• address = "Hangzhou, China",
• month = "8-12 " # jun,
• publisher = "IEEE",
• keywords = "genetic algorithms, genetic programming, Integrated circuits, Training, Cloud computing, Sequential analysis, Schedules, Processor scheduling, Dynamic scheduling, Dispatching, Resource management, EDA workflow scheduling, multi-objective multi-tree genetic programming, cloud computing resource management",
• isbn13 = "979-8-3315-3432-5",
• timestamp = "Mon, 30 Jun 2025 01:00:00 +0200",
• biburl = "https://dblp.org/rec/conf/cec/SunM25.bib",
• bibsource = "dblp computer science bibliography, https://dblp.org",
• URL = "https://doi.org/10.1109/CEC65147.2025.11043076",
• DOI = "10.1109/CEC65147.2025.11043076",
• abstract = "As the complexity of integrated circuit (IC) design continues to increase, electronic design automation (EDA) tools play an increasingly critical role in the chip design process. Cloud computing platforms offer novel solutions for computationally intensive EDA workflows. However, with the growing number of EDA workflows submitted to cloud computing platforms, how to efficiently schedule these continuously arriving workflows to maximize resource use has become a significant research challenge. Existing studies predominantly rely on manually designed dispatching rules, which lack automation and adaptability. This paper aims to develop an automated method for generating dispatching rules to optimise task sequencing and machine assignment in EDA workflows, reducing both makespan and flow time. To achieve this, we first constructed a mathematical model and simulation environment for the dynamic scheduling problem of EDA workflows. We then employed the Multi-Tree Genetic Programming (MTGP) approach to separately generate the Task Sequencing Rule (TSR) and the Machine Assignment Rule (MAR). Experimental results on test datasets of varying scales demonstrate that, compared to manually designed heuristic rules, the proposed approach achieves significant improvements in both makespan and flow time. Additionally, our approach generates a set of Pareto-optimal solutions, each offering a unique trade-off between minimizing makespan and flow time. These solutions, for which no alternative is better in both objectives, provide decision-makers with flexible options to select the scheduling strategy that best meets their specific needs.",
• notes = "also known as \cite{sun:2025:CEC} \cite{11043076}",

}

Genetic Programming entries for Yu Sun Ying Meng

Citations