Abstract: In view of the fact that most industrial and mining enterprises are in digital transformation at present, the digitized safety production monitoring data has the characteristics of strong real-time and huge data volume. The data volume of some systems, such as steel wire rope flaw detection system, is up to several hundred thousand pieces per second. The commonly used SQL database has been unable to cope with such a huge data storage and analysis task. A high-density data acquisition system with high concurrency and high density data processing capability based on NoSQL is constructed through NoSQL database, load balancing, multithreading, SOCKET and other technical systems. The system realizes the high-density access of safety production related data of industrial and mining enterprises, and meets the hard demand of high-density data sampling cycle.