In recent years, IoT (Internet of Things) has become a trend, and various sensor devices generate large amounts of data. HBase is one of the data stores for managing such big data. We evaluated the performance of HBase for managing the power consumption data collected from 10 million smart meters. With this test, we got the following results.
- HBase showed high data insert performance by processing at least 64 requests in parallel.
- The acquisition request for time series data showed more than 35 times percormance compared with non - time series data.
- By choosing an appropriate compression algorithm, the storage size of time series data could be compressed to 1/5.
From these results, we confirmed that HBase is suitable for storing and referring data generated by many sensors. In this session, I will introduce various know-how gained through the performance test.
