摘要
越来越多的可穿戴传感器用于监测身体各种重要指标,但大规模健康监测数据的存储、管理和利用仍面临很多问题。为解决传统健康监测信息不能有效采集和存储从而形成数据孤岛等问题,基于Kafka的发布—订阅模式,实现健康监测大数据采集、传输、发布共享架构,使用HBase组件对数据进行持久化存储。通过Kafka集群测试以及建立HBase二级索引优化平台性能。实验结果表明,平台吞吐率达到27MB/s左右,查询效率比二级索引建立前提高近3倍,能有效满足健康监测大数据平台发展需要。
More and more wearable sensors are used to monitor various important indicators of the body,providing unprecedented op⁃portunities for individual real-time monitoring.However,there are still many problems in the storage,management and utilization of large-scale health monitoring data.In order to solve the problems that traditional health monitoring information cannot be effectively collected and stored,and data islands are formed,based on Kafka’s publish-subscribe model,health monitoring big data collection,transmission,and publishing shared architecture are realised,and HBase components for persistent storage of data are used.Through Kafka cluster testing and the establishment of HBase secondary indexes to optimize the platform performance,the experimental results show that the platform throughput reaches about 27MB/s,and the query efficiency is nearly 3 times higher than before the secondary in⁃dexes are established,which can effectively meet the the development needs of health monitoring big data platform.
作者
王勇
张跃
WANG Yong;ZHANG Yue(College of Information,Beijing University of Technology,Beijing 100124,China)
出处
《软件导刊》
2021年第4期188-193,共6页
Software Guide
关键词
健康监测大数据
采集传输
存储
发布共享
big data for health monitoring
collection and transmission
storage
publication and sharing
