超大数据量的快速排序法

QUICK SORT METHOD OF SUPER VAST DATA AMOUNT

在超大数据量情况下 ,传统的排序方法及分组优化法由于其自身的算法限制 ,难以更有效地提高排序效率。为此 ,作者在本文中提出了一种快速排序法 ,该方法的基本思路是用特殊设定的不同分界值把整个数据区域分解成尽可能多的小单元 ,每个数据通过与不同分界值的对比迅速归位于相应的单元内 ,然后对各单元内的数据进行排序 ,按序连接各单元 ,最终达到排序的目的。与分组优化法的计算工作量大小对比及试算结果 ,充分表明了该方法的算法可行 ,计算量少 ,排序速度更快等优点 ,特别适用于超大数据量情况的数据排序。

In the case of vast data amounts, it is difficult to improve the data sorting efficiency of tranditional sorting methods and grouping optimization methods due to their algorithm limits.A quickly sorting method has been presented in this paper, in which the whole data range are divided into several small cells as many as possible by special dividing grade enactment, the data in each cell are collected by comparison between data value and different dividing grade, and the whole data sorting procedure is completed by connecting all cells in order. By comparison with conventional methods, the new algorithm of quickly sorting data is characted by high caculating speed and it is feasible for the super vast amount of data.