Resource integrity-aware flexible resource scaling approach over sensor-cloud

Abstract

Massive internet of things (IoT) framework deployments increase edge devices usage and dependently increase the generation of data. The traditional elastic asset scheduling approach is phenomenally suitable to a single cloud environment. The prognosticative asset demand is not sufficient. The existing methods are neglecting billing mechanisms to scale up and down the asset scheduling actions. Consequently, we propose an adaptive workload prediction algorithm to schedule the resource and asset migration algorithm to accomplish low leased costs. The predictive model ensures assets scheduling at cluster-edge to reduce the latency. The migration algorithm regulates data reliability with moderate workload balancing. The simulation results exhibit an adaptive system performance such as leased cost curb, essential data integrity, and workload balancing.