Design Of Multi Source Data Flow Monitoring Platform Based On Websoket

The expansion of the earthquake monitoring and early warning network has followed a phenomenon that the amount of seismic observation data has increased sharply.Compared with the distributed big data platform to solve the problem of massive data processing,the seismic data processing software under the traditional stand-alone platform cannot meet the needs of today's earthquakes.Forecasting work,such as deformation monitoring that relies on real-time data flow monitoring,and seismic data flow response pick-up and display,etc.,are highly time-dependent monitoring work that cannot meet current needs under traditional monitoring platforms.The "polling" mode adopted by traditional monitoring platforms for data streams often leads to large server network consumption,and data processing is overly dependent on server performance and low efficiency.At the same time,multi-base-station real-time data flow visualization processing process multi-base-station data stream and multithread parallel Under execution,there will be problems such as thread blocking and waiting,resulting in missing data stream time segments.In order to meet the needs of real-time monitoring of massive data streams under current needs,this research uses B/S architecture to build a system framework based on the latest HTML5 standards,introduces multi-source data stream interfaces as server communication means,and uses Websocket technology to solve the full-duplex communication of the front and back ends.And use the hash table to optimize multithread management to ensure the complete reception of streaming data.On the client side,Echarts is used as a rendering component to realize the dynamic display of streaming data.The overall design and development of a set of multi-source data streaming real-time is based on the module embedded idea.Monitoring System.The system uses Websocket full-duplex communication technology to build a single network request and continuous data push function between the server and the client in a low network load environment;through the hash table,the complex thread switching operation under the concurrent execution of multiple threads is optimized to a single thread The thread opening and closing operations are executed sequentially,avoiding the lack of data flow time sequence caused by the blocking and waiting mechanism in multi-thread switching,and ensuring the thread atomicity between the base station data flow multi-threads;using Echarts statistical graphs for high compatibility and streaming Loading features,combined with the Web Socket protocol to achieve real-time visual chart monitoring of data flow;at the same time,the module embedded design idea also ensures the scalability of the system.The system pressure test and operation results show that the system server receives complete data stream for 113 GNSS base stations in a 12-hour running time under a high concurrency environment with a concurrent number of 60,156 parallel executions are stable,no thread blocking,front and back ends In the transmission of 690 communication data,it only took 39 ms.The server network load is low,the communication is efficient,and the monitoring results are accurate.Different from the high network load caused by data stream processing under traditional polling technology and many key problems such as the loss of data stream caused by switching blocking and waiting among multiple threads,this system constructs a Websocket channel to achieve a low-load server environment,and the system The overall interface is simple and beautiful,and the combination of hash table and multi-thread management also makes the time segment of the data received by the system complete during the thread switching process of each base station's data stream.The system not only guarantees the scalability in the future monitoring network expansion environment,but also satisfies the visualization needs of the massive seismic data stream in the future big data era,and further improves the real-time transmission and display capabilities of the data stream.