Adaptive resource management for spot workers in cloud computing environment
Lung Pin Chen, Fang Yie Leu, Hsin Ta Chiao, Hung Jr Shiu · International Journal of Web and Grid Services · 2022
Due to flexible scheduling requirements of various service applications, a cloud platform usually has some temporarily unleased machines. To make cost-effective of the platform, such a considerable number of idle workers can be collected to perform malleable tasks. However, these workers are considered unstable since they can be interrupted unexpectedly by the resource broker. This paper proposes a resource management approach that employs replication to increase resource availability. We will show that increasing the replication factor can improve the worker reliability, but on the contrary, it also worsens the overhead of computational redundancy. An algorithm that can effectively control the replication factor so as to adapt to the changing workload and maintain the system performance is also developed.