Today, data is gradually changing in popular culture in a continuous, rapid and time-varying direction. According to the new application framework, people put forward higher requirements for the performance and functions of data systems, and also put forward higher requirements for the practice and understanding of data processing. This paper aims to study the processing method of structured data and the construction of distributed processing system. This paper first analyzes the research status of distributed data systems at home and abroad, then designs a typical distributed data processing system and discusses two basic technologies of the system, and finally applies it to the project of postal centralized delivery system. This paper summarizes the structural and linguistic features of the text. According to these characteristics, this paper proposes an organizational structure. The method has three main parts: text prioritization, new text discovery and information extraction. This paper analyzes the characteristics and memory space allocation of continuous query, and designs a dynamic window query based on the greedy principle, which improves the efficiency of continuous query processing. Proven by experiment. The new word rate with a word length of more than 5 is 0%. In the case of different data volumes, the load change rate of the system is less than 0.05, and the system is stable.