| Sort algorithm is one of the fundamental techniques in computer science. Sort algorithms with high speed and small additional space are pursued by many scientists. There are A lot of compare-based and exchanging sort algorithms, for example bubble sort, select sort and the like, which have low efficient. In these type sort the best time complexity is O(NlogN) and Quick Sort[3] is the best compare-based algorithm.For handling large data and improving speed of sort, a lot of algorithms were public via researching by many persons, for example SIS,FlashSort,Proportion Split Sort and so on.This paper introduces a new sort algorithm, bytes-split index sort (BSIS), on an approach of non compare-based sorting. The algorithm splits every digit of a sequence into m (m>= 2)bytes and constructs m subsequences, then sorts every subsequence via using index technique which is similar to self-indexed sort.The sort algorithm---BSIS is none of business with data type and distribution of sorting data. It solved the problem that SIS sort algorithm just support positive integer and will need much memory when maximum value of sorting data is large. And avoid the problem that FLASH sort algorithm depends on distribution of sorting data.At last part of the thesis, a summary descript merits and shortcoming of the algorithm. |