Serverless computing and container-based architectures have emerged as leading paradigms for cloud-native applications, each with distinct scaling behaviors. This paper examines the scalability tradeoffs of serverless Function-as-a-Service (FaaS) platforms and containerized (non-serverless) distributed systems, providing a side-by-side comparison with traditional distributed architectures. We identify key factors impacting scalability – including latency under load, resource management overhead, cost efficiency, cold start delays, and horizontal scaling limits – and investigate how these manifest in serverless vs. non-serverless environments. Our analysis explores insights from existing literature. Our analysis shows that serverless platforms offer rapid auto-scaling and fine-grained resource usage, excelling under bursty workloads, but face challenges like cold start latency and execution time limits. Containerized microservices, in contrast, provide consistent performance for steady loads and greater control over resources, yet require complex orchestration to scale and may incur higher costs when underutilized. We present evidence demonstrating these trade-offs and discuss implications for system design. Contributions of this work include a comprehensive review of scalability challenges in modern cloud systems, a comparative performance evaluation of container orchestration vs. FaaS under varying loads, and guidelines for selecting the appropriate architecture. We conclude by outlining future research directions to mitigate current limitations (e.g., cold starts) and to explore hybrid models that blend serverless flexibility with the control of traditional distributed systems.