一种改进的冰山立方体计算方法及其在机票结算数据中的应用

Improved iceberg computation algorithm and its application in data of air ticket settlement

为了提高冰山立方体的计算性能,提出一种基于位图索引改进的DPBUC_BI(dynamic pruning based BUC_BI)算法。该算法利用位图索引按列组织的特性重新定义BUC(bottom-up computation)算法的分组操作,加快了数据的加载和查询;通过使用逻辑位运算实现聚合计算,提高了算法的计算性能。针对部分数据聚集现象增加动态剪枝策略,在保证算法正确性的情况下进一步提高了冰山立方体计算性能。最后将DPBUC_BI算法应用于机票结算数据的冰山立方体计算中,实验结果表明:该算法可以很好地提升计算性能,相对于经典BUC算法在时间性能上有一定的提高。

In order to improve the performance of iceberg cube computation,this paper proposed dynamic pruning based bottom-up computation with bitmap index（ DPBUC_BI） algorithm. It used the feature of bitmap index like organizing the data by columns to redefine the method of partition in the bottom-up computation（ BUC） algorithm,which accelerated the loading and query of data. The performance of the computation was improved with the aggregation computation implemented by logical bitwise computation. Aiming at the problem that a large number of data focused on several dimensionalities,it used dynamic pruning strategy in the DPBUC_BI algorithm,which optimized the performance of algorithm within the similar computational accuracy. The experimental result shows that the performance of iceberg computation in the data of air ticket settlement has good computational performance and less time computation in DPBUC_BI algorithm than that in the traditional BUC algorithm.