Review and Comparative Study of Bitmap Indexing Techniques

Decision support systems that access data from large databases are mainly designed to handle complex and ad hoc queries. Now a days, the massive, non-volatile, and subject-oriented databases can include the processing of analytical and interactive queries that need quick response time with high accuracy. To enhance the data mining queries performance, many techniques such as various types of indices, materialized views and data fragmentation are used. The bitmap indices are mostly suitable in read mostly datasets like data warehouses and transactional databases. The main benefit of using bitmap indices is that bitmap vectors can be directly accessed without decompression and it helps to improve processing time for complex and interactive queries. They significantly use low cost Boolean operations and check predicate conditions on the index level prior to accessing to the primary source data. This paper presents bitmap indexing techniques for data warehouses along with their analysis and comparison among them.