Multidimensional Data Index Structure Cloud Environment

Cloud computing as a new computing infrastructure, need a lot of follow-up work. The key of building a cloud infrastructure is to build a data storage system extensibility, and these data storage system must support a lot of workload. Existing mainstream of cloud storage system mainly include Google’s BigTable, Amazon’s chateau,etc. But most of the current cloud storage system, mostly adopts DHT (distributed hash table) way to build the data index, that is a pure Key-Value model to organize data, and can through primary Key and efficient access to data.However this type of solution is usually used after specialized processing, or based on semantic is the local mapping mechanism to optimize query performance, the solution in terms of range query and multidimensional query support have not been very ideal. So this paper tries to put forward a new kind of distributed multidimensional data in a cloud environment index structure SUB-index, put forward some reference to solve such problems.This paper tries to put forward the distributed multi-dimensional data in a cloud environment index structure SUB-index, the UB tree structure is used to as a multidimensional data index research base, try to use a similar linear structure to construct a distributed index, not only combines the skip-webs architecture doesn’t need to balance adjustment feature, to replace complex adjustment in balanced tree, but also consider the storage structure and disk structure corresponds to. Try to put forward effective multidimensional data in a cloud environment index system as a new reference. In SUB-index of multi-dimensional data indexes based on the architecture, design and related query algorithms, including query and range query, and update the insertion and deletion algorithm. Paper also puts forward the distributed multi-dimensional data indexes of the complexity of the query and range query and to consider the factors in the theory, design the related testing. As can be seen from the result of the experiment, SUB-index index structure is effective, and provides a good support in terms of range queries.