The Cloud-Based Sensor Data Warehouse
暂无分享,去创建一个
As advances in technology, sensors have been widely applied to various different fields. Sensor data with time characteristic, therefore after a long period of observation that demand for data storage and analysis are very large. The relational database is the most widely used database architecture that through the normalization process to design the table structure. The tables related to each other through the foreign key fields for data link. The greatest benefit of relational database is easy to manage data. The data after normalization can be avoided through the problem of inconsistent data. But the drawback is that data is stored using row-oriented which all rows in a table are the same. When the data is increasing exponentially such as the need to change table structure it takes more time to consume to restructure data. And, the traditional relation database often faces with the need to expand storage space problem. The most often way to solve this problem is to extend the storage space vertically, however, all data stored on a single server will cause a larger workload on the server.
[1] Wilson C. Hsieh,et al. Bigtable: A Distributed Storage System for Structured Data , 2006, TOCS.
[2] D. Carstoiu,et al. Hadoop Hbase-0.20.2 performance evaluation , 2010, 4th International Conference on New Trends in Information Science and Service Science.