Blockchain Based Delay-Tolerant Resource Optimization in Fog and Cloud Layers Utilizing NNGOA and LS2BiOLSTM

Abstract

Resource Optimization (RO) in fog and cloud layers enhances performance, minimizes costs, and ensures seamless integration of distributed systems. However, prevailing works failed to perform resource optimization in both fog and cloud layers due to their complex and disparate architectures. Therefore, the proposed work performs resource optimization efficiently in both fog and cloud layers by predicting the network traffic congestion using Neuron Northern Goshawk Optimization Algorithm (NNGOA) and Log Sigmoid Softplus Bidirectional Orthogonal Long Short-Term Memory (LS²BiOLSTM). At first, the Cloud Users are registered and logged in for task assignments. Meanwhile, the Smart Contract (SC) based Service Level Management (SLM) is created for tasks. After that, the signature is created for SLA and is verified during task assignment. For predicting the network traffic congestion in tasks, LS²BiOLSTM is utilized. Then, the predicted congestion tasks are clustered and mapped into a fog layer. Simultaneously, from the Cloud Server (CS), the data center is prioritized using SoftSign Bell-Fuzzy (SSB-Fuzzy). Finally, the resources are optimized efficiently with a high accuracy of 98.1259% using NNGOA, which outperforms the existing methodologies.