Please use this identifier to cite or link to this item:
http://dspace.bits-pilani.ac.in:8080/jspui/xmlui/handle/123456789/8132
Title: | New binning strategy for bitmap indices on high cardinality attributes |
Authors: | Goyal, Navneet Sharma, Yashvardhan |
Keywords: | Computer Science Bitmap indices |
Issue Date: | Jan-2009 |
Publisher: | ACM Digital Library |
Abstract: | Bitmap indices are the preferred indexing structures for read only & high dimensional data in data warehouses and scientific databases. High cardinality attributes pose a new challenge in terms of having space efficient bitmap indices. Binning is a common technique for reducing space requirements of bitmap indices. It is found that binning has an adverse affect on the query performance. A new efficient binning strategy is proposed for bitmap indices for high cardinality attributes. Exact bins are created based on query distribution. Exact bins are allowed to overlap. This gives a considerable performance advantage over the conventional non-overlapping bins at the expense of marginal increase in space overheads. Overlapping bins minimize the number of candidate-checks that need to be performed for a given set of queries. Algorithms are also presented for performing candidate checks more efficiently. Experimental results are presented in support of the new binning strategy. |
URI: | https://dl.acm.org/doi/10.1145/1517303.1517327 http://dspace.bits-pilani.ac.in:8080/xmlui/handle/123456789/8132 |
Appears in Collections: | Department of Computer Science and Information Systems |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.