A new research paper proposes IoTGA-SRC², a genetic algorithm variant that optimizes task offloading decisions across IoT devices, fog nodes, and cloud servers while respecting strict deadlines. Unlike traditional approaches that focus solely on minimizing latency or energy, IoTGA-SRC² incorporates deadline constraints directly into the fitness function, enabling more realistic scheduling for time-sensitive IoT workloads. The algorithm uses a novel chromosome encoding scheme that represents both task assignment and resource allocation, and employs specialized crossover and mutation operators to maintain feasibility. Early results show significant improvements in deadline satisfaction rates compared to baseline heuristics, particularly under high system load. This work is relevant for developers building edge-cloud orchestration systems for industrial IoT, smart cities, and autonomous systems where missed deadlines can have serious consequences. The paper's approach could inspire practical scheduling libraries or middleware for IoT platforms.
This paper introduces IoTGA-SRC², a genetic algorithm designed for deadline-aware task offloading in IoT-edge-cloud environments. It addresses the challenge of balancing local computation, fog node processing, and cloud execution while meeting task deadlines. The approach is significant for real-time IoT applications where latency and resource constraints are critical.