Low-latency processing is necessary for real-time applications due to rapid IoT progress. Traditional cloud-only infrastructure often struggles to meet this requirement. Fog computing overcomes this limitation by bringing processing power closer to the data source, enabling quick handling of latency-sensitive tasks. Integrating microservices into fog environments improves system modularity, resource utilization, and fault tolerance, particularly in applications like autonomous systems, smart cities, and healthcare. This study explored fog-based microservices, focusing on scheduling, resource allocation, and microservices placement strategies in fog-cloud ecosystems. Key challenges include achieving fault tolerance, optimizing resource-constrained environments, and securing distributed data across fog nodes. The study compares performances of microservices placement techniques such as Particle Swarm, Genetic Algorithm (GA), and Greedy Placement, based on reliability, cost, resource efficiency, latency, and Particle Swarm Optimization (PSO). Results show PSO outperforms alternative strategies by effectively distributing microservices across fog and edge layers, reducing latency and enhancing system resilience.