Abstract:
In the rapidly evolving field of smart networking, real-time data processing is critical for efficient system performance. Currently, conventional methods such as Round-Robin scheduling, Static Resource Allocation, and Shortest Job First (SJF) scheduling are widely used to distribute workloads. However, these methods often fall short in dynamic environments where data flow and network demands are unpredictable, leading to inefficiencies and increased latency. Research gaps in existing systems primarily include their inability to adapt to changing network conditions and their poor scalability under varying loads. These drawbacks highlight the need for a more flexible and responsive approach. This paper introduces the Dynamic Adaptive Workload Distribution Algorithm (DAWDA), a novel method designed to address these limitations. DAWDA dynamically adjusts resource allocation based on real-time network data and workload characteristics, ensuring optimal performance and minimal response times. The proposed method leverages advanced machine learning techniques, including Reinforcement Learning and Predictive Modeling, to anticipate network demands and adjust resources preemptively. In testing, DAWDA demonstrated a 0.30% increase in throughput, a 0.25% reduction in latency, and a 0.20% improvement in resource utilization, significantly outperforming traditional methods such as Round-Robin scheduling, Static Resource Allocation, and Shortest Job First scheduling. Overall, DAWDA not only resolves the inefficiencies found in existing systems but also sets a new standard for workload distribution in smart networking environments, promising substantial improvements in real-time data processing capabilities.
