Fast and Cost-Effective Load Balancing Method for Range Queriable Cloud Storage

The rising popularity of big data processing of social networks and sensor networks has made the range queriable cloud storage increasingly important. To support range queries, the data locality is preserved strictly, which makes the load balancing among nodes a challenging task. Currently, most of the range queriable cloud storage systems adopt the centralized load balancing method, which is inadequate in a large-scale, failure-prone environment. In this work, a fully distributed load balancing method is presented. In our method, each physical node is divided into multiple virtual nodes with equal size, and all the virtual nodes are organized with range queriable P2P. Load balancing is conducted in a lower level (among virtual nodes) and a higher level (among physical nodes) simultaneously with only local knowledge. Our method is expected to achieve similar performance as the centralized method with relatively low cost.