Edge clouds face challenges of resource assignment and load balancing due to variability of user location (mobility), server load and network state. Dynamic resource migration techniques are considered necessary to achieve load balance, fault tolerance and system maintenance objectives. Container migration is emerging as a potential solution that enables dynamic resource migration in virtualized networks and mobile edge cloud (MEC) systems. This paper proposes a traffic aware container migration approach and validates it with an end-to-end system implementation using a pure container hypervisor called LXD (Linux Container Hypervisor). The container migration model is then evaluated for real-time applications such as license plate recognition running in a mobile edge cloud scenario based on city-scale mobility traces from taxicabs in San Francisco. The system evaluation considers key metrics associated with application quality-of-experience (QoE) and network efficiency such as the average system response time and the migration cost for different combinations of load, compute resources, interedge cloud bandwidth, network and user latency. A specific compute resource and network-aware distributed resource migration algorithm called ‖ShareOn‖ is proposed and compared with alternative techniques using the San Francisco MEC model.