With the rapid advancement of digitization and intelligence, enterprise big data processing platforms have become increasingly important in data management. However, traditional monolithic architectures, due to their high coupling, are unable to cope with increasingly complex demands in the face of business expansion and increased data volume, resulting in limited platform scalability and decreased data collection efficiency. This article proposes a solution for enterprise big data processing platform based on microservice architecture, based on the concept of Domain Driven Design (DDD). Through in-depth analysis of business requirements, the functional and non functional requirements of the platform in various scenarios were determined, and the DDD method was used to decompose the core business logic into independent microservice modules, enabling data collection, parsing, cleaning, and visualization functions to be independently developed, deployed, and upgraded, thereby improving the flexibility and scalability of the system. This article also designs an automated data collection process based on microservices and proposes an improved dynamic scheduling algorithm to efficiently allocate data collection tasks to Docker nodes, and monitor the collection progress and service status in real time to ensure the accuracy and efficiency of data collection. Through the implementation and testing of the platform, it has been verified that the enterprise big data processing platform based on microservice architecture has significantly improved scalability, data quality, and collection efficiency.