Affiliation:
1. The Univ. of Texas, Austin
Abstract
Many data-parallel algorithms—Fast Fourier Transform, Batcher's sorting schemes, and the prefix-sum—exhibit recursive structure. We propose a data structure called
powerlist
that permits succinct descriptions of such algorithms, highlighting the roles of both parallelism and recursion. Simple algebraic properties of this data structure can be explotied to derive properties of these algorithms and to establish equivalence of different algorithms that solve the same problem.
Publisher
Association for Computing Machinery (ACM)
Cited by
62 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Exploring Scalable Parallelization for Edit Distance-Based Motif Search;IEEE/ACM Transactions on Computational Biology and Bioinformatics;2023-03-01
2. On Generalizing Divide and Conquer Parallel Programming Pattern;Mathematics;2022-10-23
3. Reflections on the Design of Parallel Programming Frameworks;Communications in Computer and Information Science;2021
4. PList-based Divide and Conquer Parallel Programming;Journal of communications software and systems;2020-05-27
5. Enhancing Java Streams API with PowerList Computation;2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW);2020-05