Deadline and budget-constrained archimedes optimization algorithm for workflow scheduling in cloud

Abstract

Cloud computing has revolutionized various domains over the past decade, providing accessible computational and storage resources at reduced costs. The exponential growth in data volumes and processing complexity, particularly due to the proliferation of IoT devices and applications across fields such as business, education, and agriculture, requires scalable computing resources and efficient processing. Workflow scheduling in cloud computing, an NP-hard optimization problem, involves allocating resources to tasks within a workflow and determining their execution sequence. Despite numerous heuristic, metaheuristic, and hybrid approaches, there remains a need for scheduling algorithms with lesser computational complexity to optimize makespan and cost efficiency, as well as ensure SLA compliance. This paper introduces a novel multi-objective metaheuristic solution, the Deadline and Budget constrained Archimedes Optimization Algorithm (ADB), which addresses workflow scheduling by optimizing makespan and cost while adhering to deadline and budget constraints. Extensive experiments on a well-known cloud simulation tool, Workflowsim, using scientific workflows demonstrate significant improvements in makespan (20%), cost (5%), resource utilization (15%), and energy consumption (9%). Performance observations on Pareto optimality metrics show that our approach has a higher hypervolume for 80% cases, it dominates state of the art by at least 83%, and the s-metric value of our approach is lower for 95% cases, alongside statistical validation using t-tests and ANOVA, confirming the efficacy of our method compared to state-of-the-art approaches. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.