Research And Design Of Healthcare Data Management System Based On Hadoop

With the development of new technologies such as mobile Internet,Internet of Things,cloud computing and wearable equipment,and the proposition of the health information service and the wisdom medical service for the whole people,the application of healthy large data is promoted.In the face of massive medical data,how to efficiently store and quickly process data becomes the current major demand.During processing huge amounts of medical data,the traditional database is facing many problems,such as storage overwhelmed,high cost,computing power can not match.In view of these problems,the Hadoop architecture can bring about the distributed storage and processing of massive data,and combining the clustering characteristics,the low cost,high efficiency and scalable data processing is achieved.Based on the characteristics of medical and health data,this paper puts forward the research of Hadoop-based medical health data management system,which is divided into two modules:Android-based medical health management system and Hadoop-based medical health data management system.Based on the Android health management system and the Android smart phone,the mobile management of Internet APP is designed.The software includes user registration registration,heart rate detection,weight measurement,motion measurement,data statistics and health recommendations.The software can achieve the measurement of physiological parameters,improve motion algorithm,analyze the measurement results,and provide health advice and other functions.Hadoop-based healthcare data management system is to use new technology Hadoop to store and manage data quickly.In this paper,we set up a cluster center of four machines,and then taking into account the fact that medical data requires frequent real-time write read characteristics,we set up HBase database in Hadoop cluster center,replace HDFS,and re-design the data format to achieve HBase storage medical data.Then,we use the powerful parallel computing model,Hadoop’s MapReduce,to analyze healthcare data and complete the management and storage of medical and health data.Improve the multi-conditional query of HBase,combined with HBase and Solr,the easy multi-conditional query is achieved and the query efficiency is improved in the end.Finally,the system performance and reliability test of the client and server of the medical and health data management system are designed and simulated.The verification results show that the system can meet the functional design and reliability of the demand.Android various functional modules test qualified,MapReduce performance test and HBase database write test is consistent with the expected,HBase and Solr combined query data time is greatly reduced.Therefore,Hadoop-based health care data management system can better meet the medical and health data requirements.