Bitmap indexing a suitable approach for data warehouse design
Bitmap indexing a suitable approach for data warehouse design
复制标题
位图索引是数据仓库设计的合适方法
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
K. Sarika
中科院分区：
文献类型：
--
作者：
K. Sarika
-数据仓库是面向主体、集成化、时变、非易失性的庞大数据库的集合。由于它是一组庞大的数据库，快速的数据访问是任何数据仓库的主要性能参数。通常，根据组织的某些决策过程的需要，对从数据仓库检索到的信息进行汇总或聚合。要检索这样的信息，要触发的查询本质上是聚合函数，后面跟着有子句。有效地从数据仓库中提取信息是摆在研究人员面前的难题。由于它是一个巨大的数据库，与普通数据库相比，访问信息所需的时间更多。因此，在这个庞大的数据库上创建索引是必不可少的，这对于提高数据仓库的性能非常重要。选择合适的索引可以减少查询的执行时间，提高数据仓库的性能。目前，b树索引被广泛应用于不同的数据库产品中。对于在任何表上创建索引，他们使用B树方法。B树索引对于需要频繁更新的数据库很有用，比如联机事务处理系统（OLTP）。对于数据仓库和联机分析系统（OLAP）来说，这是一种耗时的方法。数据仓库不经常更新，因此位图索引是合适的选择。我们必须在开始时仅在所需向量上创建位图索引。一旦它被创建在固定的数据库上，我们就可以在任何时候使用它进行任何查询。根据查询的要求，选择位图并执行查询。位图索引是数据仓库的合适选择，因为它具有非易失性和庞大的数据集等特点。
— Data warehouse is a collection of huge database which is subject oriented, integrated, time-variant and non volatile. As it is a set of huge database, fast data access is the major performance parameter of any data warehouse. Generally the information retrieved from Data Warehouse is summarized or aggregated as it is required for some decision making process of organization. To retrieve such a information queries to be fired is of the nature aggregation function followed by having clause. Extracting information efficiently from data warehouse is the challenge in front of researchers. As it is a huge database time required to access information is more compare to normal databases. Due to this creating index on this huge database is essential and it is important for increasing the performance of data warehouse .Selection of appropriate indexing decreases the query execution time and the performance of data warehouse is increase. Presently B-tree indexing is used in different database products. For creating the index on any table they uses B Tree approach. B tree indexing is useful for the databases where the frequent updates are required like On Line Transaction Processing system(OLTP). It is a time consuming approach for data warehouse and On Line Analytical System(OLAP). Data warehouse is not frequently updated so Bitmap indexing is appropriate choice for the same. We have to create bitmap index on required vector at the start only. Once it is created on fixed database we can use it any time for any query. As per the requirement of query we have to select bitmap and execute query. The bitmap indexing is appropriate choice for Data warehouse only because of its feature like it is non volatile and huge data set .