A lower bound for distributed averaging algorithms
Optimization and Control
2015-03-13 v1
Abstract
We derive lower bounds on the convergence speed of a widely used class of distributed averaging algorithms. In particular, we prove that any distributed averaging algorithm whose state consists of a single real number and whose (possibly nonlinear) update function satisfies a natural smoothness condition has a worst case running time of at least on the order of on a network of nodes. Our results suggest that increased memory or expansion of the state space is crucial for improving the running times of distributed averaging algorithms.
Cite
@article{arxiv.1003.5941,
title = {A lower bound for distributed averaging algorithms},
author = {Alex Olshevsky and John N. Tsitsiklis},
journal= {arXiv preprint arXiv:1003.5941},
year = {2015}
}