A multi-objective crow search algorithm for optimizing makespan and costs in scientific cloud workflows (CSAMOMC)

Abstract

Nowadays, with the rapid expansion of cloud computing technology in processing Internet of Things (IoT) workloads, the demand for data centers has significantly increased, leading to a surge in CO₂ emissions, power consumption, and global warming. As a result, extensive research and initiatives have been undertaken to tackle this problem. Two specific approaches focus on enhancing workload scheduling, a complex problem known as NP-Hard, and integrating scheduling into scientific workflows. In this investigation, we present a multi-objective Crow Search Algorithm (CSA) for optimizing both makespan and costs in scientific cloud workflows (CSAMOMC). We conduct a comparative analysis between our approach and the well-known HEFT and TC3pop algorithms, which are commonly used for reducing makespan and optimizing costs. Our findings demonstrate that CSAMOMC is capable of achieving an average makespan reduction of 4.42% and a cost reduction of 4.77% when compared to the aforementioned algorithms.