| Popis: |
An online distributed system which is helpful for users to contribute their surplus resources to handle the wide range of tasks is called volunteer computing. In Volunteer Computing System (VCS) technology, there exists heterogeneous devices which vary in terms of processing power, latency, cost, and energy efficiency. VCS is heterogeneous in nature so, it is essential to utilize all the resources to provide high Quality of Service (QoS), innovation and cost effectiveness for all requests. The dynamism and heterogeneity of VCS makes it essential to utilize all the resources for providing excellent Quality of Service(QoS). This innovation is also cost effective for all the requests. Task scheduling problems that are classified as NP-hard are very challenging in heterogeneous VCS environments. Therefore, two scheduling algorithms Min-CCV and Min-V are proposed for volunteer computing systems. The primary objective of the proposed algorithms is to improve network performance and also enhance the QoS by reducing the computational, communicational, and violational costs for various Internet of Things (IoT) based applications which in turn improves the quality of life. A task scheduler module allocates tasks Ti as a group of n jobs that are identified T1, T2, T3,…….,Tn for a volunteer computing system. These tasks have specific attributes like memory requirement, input and output file sizes, QoS requirements, etc. The outcome of the simulator describes that the proposed algorithm can allocate tasks in a more efficient way to the volunteer fog-cloud environment in comparison to existing methods. As compared to the generic-based algorithms, the proposed Min-CCV and Min-V algorithms enhance the deadline satisfaction rate to approximately 99.5% and decrease the costs incurred by 15 to 53%. The comprehensive simulation results give the better outcome of proposed work as compared to existing practices. |
