Measurement: Sensors (Feb 2024)
Energy-efficient resource allocation in cloud infrastructure using L3F-MGA and E-ANFIS
Abstract
In industry along with academia, Task Scheduling (TS) along with Resources Allocation (RA) remains a challenging issue even though lots of work was carried out in the Cloud Computing (CC) field. CC uses massive virtualized Data Centers (DC). Still, it took an enormous quantity of energy on account of the constant demand by users for resources in addition to inefficient RA techniques. The higher Energy Consumption (EC) because of improper RA inside DC can be overcome using energy-aware RA techniques. Therefore, this paper proposes a Levy flight Firefly Modified Genetic Algorithm (L3F-MGA) as well as an Entropy-based Adaptive Neuros-Fuzzy Inferences System (E-ANFIS) for energy-efficient Resources Allocation (EE-RA) in cloud structure that manages the user demands and also lessens the EC of Cloud Data Center (CDC). In the proposed work, centred on task features together with cloud resource features, the input tasks are arranged using L3F-MGA. Conversely, the Virtual Machine (VM) status is identified. After that, centred on the task feature and VM status details, the E-ANFIS allows the active VM to complete the tasks effectively. The experiments compare the proposed method's performance for effective RA with other associated works. The proposed RA technique shows EC by 11 % lesser than the existing CM-GA. The Processing Time of E-ANFIS is 8 % lesser than CM-GA, 14 % less than ANFIS, 20 % less than GA, and 16 % less than PSO.
