This chapter provides a systematic exposition of the core architecture underpinning big data analysis, focusing on its capacity to overcome the limitations of traditional data systems in handling large-scale, heterogeneous datasets. It introduces a layered architectural model designed to support distributed storage, computing, and querying capabilities, which collectively address challenges in scalability, fault tolerance, and processing efficiency. Central to this architecture are several widely adopted frameworks: Apache Hadoop, a batch-oriented processing engine based on the MapReduce paradigm; Apache Spark, a unified analytics engine that supports both batch and stream processing with in-memory computing; Hadoop Distributed File System (HDFS), which ensures high fault tolerance and scalability for data storage; Apache HBase, a non-relational, column-family NoSQL database optimized for sparse data access; and Apache Hive, a data warehousing tool that enables SQL-like querying over distributed data. Through examining the design principles, functionalities, and application contexts of these frameworks, the chapter elucidates how they collectively constitute a robust and scalable ecosystem for modern big data analytics.