Fault Tolerance Based Indexing for Multidimensional Data Bases

摘要

Multidimensional databases are main source dataset for data analytics. They are designed to provide fast and efficient backend support. Efficient physical layer design of these data bases is foremost requirement. Efficiency of any storage system depends on the indexing system, which is measured in terms of time and space complexity of the indexing system. In this paper new dimension of accuracy is proposed which along with time and space which will be measure of efficiency for indexing system at physical level of multidimensional data base. Fault Tolerance capability of self-organizing neural network is used for mapping data at conceptual level to physical storage level. Results of the research are very exciting and can be used for further development of the similar techniques.