Efficient Parallel Binary Search on Sorted Arrays

Efficient Parallel Binary Search on Sorted Arrays
复制标题

排序数组上的高效并行二分搜索

DOI:
--
复制
发表时间:
1990
期刊:
--
影响因子:
--
通讯作者:
D. Chen
D. Chen
中科院分区:
--
文献类型:
--
作者:
D. Chen

文献摘要

被引文献

相似文献

令A为N数字,B数字是M数的数组,其中A和B分类为N <m。 (i)bci)::; a(j)<bci + i),其中0 s i:s m(带有b(o)= -00和b(m + 1)= +00)为此,EREW-PRAM上的算法已经给出了[i,8]。使用O((nlog(mjn»jlogm)erew-pram处理器)提出一种并行算法以在O(logM)时间求解它。OUF解决方案改善了以前的已知结果,可以在时间或总工作上复杂性,并且可以是用于获取使用OEM/ lagm/ lagm)EREW-PRAM处理器在O(logM)时间中在O(logM)时间中合并两个大小m的不同平行算法FOF。
Let A be an array of n numbers and B an array of m numbers, where A and B are sorted and n < m. We consider the problem of determining for each element AU), 1 ::; j ~ n, the element B(i) such that BCi) ::; A(j) < BCi + I), where 0 S i :S m (with B(O) = -00 and B(m + 1) = +00). Efficient parallel algorithms on the EREW-PRAM for this problem have been given [I, 8]. In this paper, we present a parallel algorithm to solve it in O(logm) time using O((nlog(mjn»Jlogm) EREW-PRAM processors. OUf solution improves the previous known results either on the time or on the total work complexity, and it can be used to obtain a different parallel algorithm fOf merging two sorted arrays of size m each in O(logm) time using Oem/ lagm) EREW-PRAM processors.