This chapter provides a comprehensive overview of major computational models in the domain of big data analytics, focusing on their theoretical foundations, system architectures, and representative frameworks. It first introduces the static batch processing model, exemplified by the MapReduce paradigm, which partitions large-scale computing tasks into subtasks executed in parallel to improve throughput and efficiency. However, given the growing demand for low-latency and real-time responses, the chapter further explores the stream processing model, represented by Apache Storm, which enables real-time data processing through a distributed and customizable architecture. Additionally, it examines the graph computing model, as implemented in frameworks like Pregel, which utilizes a vertex-centric, synchronous parallel processing mechanism to efficiently handle large-scale graph-structured data. Beyond computation, the chapter also discusses the role of data visualization technologies, which translate complex analytical results into intuitive graphical formats, enhancing interpretability and user engagement. By analyzing these diverse computational models—each tailored to specific data characteristics and analytical requirements—this chapter offers a holistic perspective on the evolving landscape of big data processing.