The Internet of Things, with its promise of ubiquitous connectivity, leads to the connectivity of billions of devices continuously generating a sheer volume of data processed by Cloud-centric Internet of Things (CIoT) based architecture. Cloud data centers reside multi-hop away from the end user, resulting in certain limitations like long latency, bandwidth, and scalability. Fog computing addresses these challenges by extending cloud-computing capabilities to the edge of the network, thus alleviating these concerns. However, the efficient execution of diverse nature fog applications on these distributed, heterogeneous, and resource-constrained fog devices needs efficient resource management techniques. Among these techniques, distributed task offloading is the one that efficiently moves the tasks from resource-limited fog devices to multiple resource-rich devices. An efficient distributed task-offloading algorithm is imperative to minimize latency and cost, optimize resource utilization, conserve bandwidth, and improve the quality of service. In this paper, we propose a multi-objective optimization-based distributed task-offloading algorithm based on NSGA-II, which reduces latency, network utilization, cost, and energy consumption. We evaluate the proposed offloading algorithm using iFogSim in comparison with PSO and FCFS approaches. The results show that the delay, cost and network usage of the proposed algorithm is much better as compared to baseline algorithms.