Abstract：Mass flow data has the characteristics of large amount,fast update speed,multi dimension and multi attributes.The storage and query of mass flow data is one of the hot topics in the academic and industrial fields in recent years.HBase system provides a set of technology and system platform for the storage and management of massive data streams.However,HBase only supports primary key index,which leads to low efficiency of no primary key data query,especially for multidimensional data.An index structure with high insertion and query efficiency is proposed for traffic flow data TA-index.Characteristics of spatial and temporal locality TA-index consider the data access to more accurately obtain data by time and attributes of different classification index,the index to reduce the amount of data,provide real-time data analysis ability.Experimental results show that the proposed algorithm is more efficient than the existing algorithms,and has high scalability,which can support high throughput and high efficiency.
2017, Vol. 38 
