Design Of Intelligent Gas Data Stream Processing System Based On Cloud Platform

The traditional gas meter has the problems of low meter reading efficiency,high labor cost,easy record error,difficult maintenance and low safety.Intelligent gas meter technology arises at the historic moment.It solves a series of problems,such as manual meter reading,gas leakage automatic alarm,gas equipment fault detection and so on.Compared with the traditional table,it is more advanced,more efficient and safer.However,compared with intelligent gas equipment,gas platforms still remain on simple multithreaded programs.The traditional gas platform processing data have little concurrency,poor scalability,low efficiency,and the simple processing of data types,such as Gao Shiyan,and so on.Therefore,the characteristics of intelligent gas equipment data are analyzed in depth.And comparing the existing mainstream technology,using the cloud platform to solve the problem of the poor scalability of the traditional platform.Storm solves the problem of the traditional gas platform Gao Shiyan and high concurrency.Java NIO2 solves the problem of a large number of data reading and writing performance under high concurrency.The traditional gas platform has the problem of poor scalability,once the gas data suddenly increase.The server can easily lose data in processing data because of insufficient performance,and the problem of growing data is only solved by replacing the server's hardware.This will lead to continuous replacement of hardware and unnecessary trouble and cost to the system.Based on this,this system builds a cloud platform to expand the increasing data volume in real time according to the characteristics of cloud platform on demand and real-time expansion.Intelligent gas data requirements are characterized by low delay,high concurrency and high reliability.Based on the characteristics of the above gas data.In this paper,the distributed flow system Storm is used to process gas data.Benchmark tests show that each node processes one million data tuples in a single second.More importantly,when dealing with data failure,it will process the failed data to another node to continue processing.Ensure that every data is handled correctly.In addition,the real-time streaming system Storm can also increase the amount of concurrency that the system can handle by expanding its nodes.In the traditional platform for processing gas data,Java uses blocking flow technology.The blocking stream must wait for the end of a data processing to process the next data,which will greatly reduce the IO performance of the system.Based on this,this system uses a new Java technology NI02 to deal with gas data,and compares the performance of the two IO flow techniques and analyses the experimental results.After the design of the system is completed,the function test and pressure test of the intelligent gas system based on real-time gas data are carried out in this paper.During the stress test,two tests were carried out from the following aspects:concurrent capability and IO performance of the system.The test results show that the system designed in this paper is sufficient to meet the requirements of the existing intelligent gas equipment data.