Abstract:

Hybrid range partitioning strategy introduces a formula to compute the amount of nodes to distribute data and the data partitioning strategy based on  identical range sizes then, an enhanced hybrid range partitioning strategy achieves data storage balancing and solves data skew between the nodes in the parallel real-tlme database system by varying range sizes. But, in order to achieve the query load balancing during the system runtime, the strategies   above should be improved. An improved hybrid range partitioning strategy introduces the definition of Heat on behalf of query load. What＇s more, it inttroduces the data partitioning strategy during system initialization, and data migration strategy during system runtime

Key words: （parallel real-time database, hybrid range partitioning strategy, data skew, query load, heat）