An efficient scientific workflow scheduling on multi-cloud environment using adaptive golden eagle optimization

Cloud computing is a model for new technologies and the ability to deliver consistent cloud services. One of the essential features of cloud computing is the provision of "unlimited" computing resources to users on demand. However, single cloud-holding resources are generally limited and may not be able to cope with the sudden surge in user needs. So, the multi-cloud concept is introduced to support resource sharing between the clouds. These days, providing assets and administrations from multiple clouds is becoming an undeniably inspiring worldview. Traditional research in scheduling in the cloud is aimed at improving the cost or makespan. Nevertheless, the reliability of work process scheduling is a fundamental concern and, surprisingly, the main measure of QoS. Therefore, in this paper, multi-objective scheduling for a logical work process in a multi-cloud environment is proposed, the point of which is to control the work process while at the same time cost and makespan while fulfilling the requirement of reliability. To achieve this concept adaptive golden eagle optimization (AGEO) algorithm is designed. The validation of the proposed algorithm takes solution encoding, fitness calculation, and updating functions. For experimental analysis, different workflow model is used and performance is evaluated using different metrics.