Parallel Meta-Heuristics for Solving Dynamic Offloading in Fog Computing

The internet of things (IoT) concept has been extremely investigated in many modern smart applications, which enable a set sensors to either process the collected data locally or send them cloud for remote processing. Unfortunately, datacenters are located far away from IoT devices, and consequently, transmission may be delayed. In this paper, we investigate fog computing, is new paradigm that overcomes issues computing. More importantly, dynamic task offloading computing challenging problem requires an optimal decision processing tasks generated each time slot. Thus, exact optimization methods based on Lyapunov function have widely used solving represents NP hard problem. To overcome scalability issue techniques, explored famous population meta-heuristics optimizing using Orthogonal Frequency Division Multiplexing (OFDM) communication. Hence, parallel multi-threading framework proposed generating solution while selecting best sub-carrier offloaded task. our contribution associates thread device generates random solutions. Next, updated evaluated according fitness considers tradeoff between delay energy consumption. Upon arrival at slot, evaluation performed maintaining some individuals previous criteria. Our results compared achieved optimization. They demonstrate convergence function, Particle Swarm Optimization (PSO) approach, performance terms offline error execution cost.