Affiliation:
1. IBM Research Division, T.J. Watson Research Center, Yorktown Heights, NY 10598, USA
Abstract
We consider a message-passing system of n processors, each of which initially holds one piece of data. The goal is to compute an associative and commutative census function f on the n distributed pieces of data and to make the result known to all processors. To perform the computation, processors communicate with each other by sending and receiving messages in specified communication rounds. We describe an optimal algorithm for this problem that requires the least number of communication rounds and that minimizes the time spent by any processor in sending and receiving messages.
Publisher
World Scientific Pub Co Pte Lt
Subject
Hardware and Architecture,Theoretical Computer Science,Software
Cited by
20 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. Uniform Algorithms for Reduce-scatter and (most) other Collectives for MPI;2023 IEEE International Conference on Cluster Computing (CLUSTER);2023-10-31
2. Fast(er) Construction of Round-optimal $n$-Block Broadcast Schedules;2022 IEEE International Conference on Cluster Computing (CLUSTER);2022-09
3. Non-clairvoyant reduction algorithms for heterogeneous platforms;Concurrency and Computation: Practice and Experience;2014-07-30
4. A Model for Learning;Fault-Tolerant Search Algorithms;2013
5. Group Testing;Fault-Tolerant Search Algorithms;2013