<p>Edge computing has come up as integral chapters in computing systems to supplement conventional cloud computing, and in some places inevitable too. Beyond data-centers, the efforts are going on to reach near the end-user as efficiently as possible. At the same time, a rethink about task allocation has been necessary to maintain the QoS (Quality of Service), especially in cases where edge servers deal with heterogeneous mobile edge devices and latency-aware tasks. This paper presents a novel task allocation algorithm for IoT-edge-cloud environments leveraging Ant Colony Optimization (ACO). Addressing the challenges of heterogeneous mobile edge devices and latency-sensitive tasks, our ACO-based approach dynamically maps tasks to data centers, minimizing load and optimizing resource utilization. Unlike traditional Trade-Off, Round Robin and Fuzzy Logic strategies, experimental results within the PureEdgeSim framework demonstrate significant improvements. Specifically, our algorithm reduces the task orchestration waiting time by approximately 20% for 500 edges compared to the other strategies, so it increases the successful task execution rate. This enhanced performance stems from the ACO’s inherent ability to efficiently balance server load, leading to reduced energy consumption and improved Quality of Service (QoS).</p>