Distributed Convergence Detection Based on Global Residual Error Under Asynchronous Iterations - Archive ouverte HAL Access content directly
Journal Articles IEEE Transactions on Parallel and Distributed Systems Year : 2018

Distributed Convergence Detection Based on Global Residual Error Under Asynchronous Iterations

Abstract

Convergence of classical parallel iterations is detected by performing a reduction operation at each iteration in order to compute a residual error relative to a potential solution vector. To efficiently run asynchronous iterations, blocking communication requests are avoided, which makes it hard to isolate and handle any global vector. While some termination protocols were proposed for asynchronous iterations, only very few of them are based on global residual computation and guarantee effective convergence. But the most effective and efficient existing solutions feature two reduction operations, which constitutes an important factor of termination delay. In this paper, we present new, non-intrusive, protocols to compute a residual error under asynchronous iterations, requiring only one reduction operation. Various communication models show that some heuristics can even be introduced and formally evaluated. Extensive experiments with up to 5,600 processor cores confirm the practical effectiveness and efficiency of our approach.
No file

Dates and versions

hal-01737234 , version 1 (19-03-2018)

Identifiers

Cite

Frédéric Magoulès, Guillaume Gbikpi-Benissan. Distributed Convergence Detection Based on Global Residual Error Under Asynchronous Iterations. IEEE Transactions on Parallel and Distributed Systems, 2018, 29 (4), pp.819 - 829. ⟨10.1109/TPDS.2017.2780856⟩. ⟨hal-01737234⟩
80 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More