An Effective Multi-Agent Ant Colony Optimization Algorithm for QoS-Aware Cloud Service Composition

Recently, service composition has gained increased attention as an auspicious paradigm to optimize the data accessibility, integrity, and interoperability of cloud computing. In this work, to solve the cloud service composition (CSC) problem, we introduce an efficient agent-based ant colony optimization (ACO) algorithm. The CSC problem aims to satisfy complex and challenging requirements of enterprises/users in a cloud environment. The challenge of such problem is the proliferation of providing similar services having similar functionality with varying quality of service (QoS) properties from different providers. Several swarm-based algorithms were introduced to solve this problem because the complexity of the problem is characterized as NP-hard, which is high. These algorithms aim to maintain a good balance between exploration and exploitation mechanisms, and to achieve this, a multi-agent based on ACO is proposed and compared with four different algorithms using 25 different real datasets. The computational results on 25 real datasets confirm the effectiveness of the multi-agent distribution of ACO process. Moreover, comparisons against the results of the four algorithms in the literature indicate that the multi-agent ACO approach is competitive with state-of-the-art algorithms.

Automated summary

In this study, an efficient agent-based ant colony optimization (ACO) algorithm is introduced to solve the cloud service composition problem, which aims to meet the complex and challenging requirements of enterprises/users in a cloud environment. The computational results demonstrate the effectiveness of the multi-agent ACO approach across 25 real datasets, showing competitiveness with state-of-the-art algorithms in literature comparisons.