# Improving all-reduce collective operations for imbalanced process arrival patterns

arXiv: https://arxiv.org/abs/1804.05349 · HTML: https://mlsanity.com/paper/1804.05349/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Apr 2018.

## Built on by (most cited first, 1 of 1)

- [Rethinking Memory and Communication Cost for Efficient Large Language Model Training](https://mlsanity.com/paper/2310.06003/) (2023)

## Abstract (arXiv)

Two new algorithms for the all-reduce operation, optimized for imbalanced process arrival patterns (PAPs) are presented: (i) sorted linear tree (SLT), (ii) pre-reduced ring (PRR) as well as a new way of on-line PAP detection, including process arrival time (PAT) estimations and their distribution between cooperating processes was introduced. The idea, pseudo-code, implementation details, benchmark for performance evaluation and a real case example for machine learning are provided. The results of the experiments were described and analyzed, showing that the proposed solution has high scalability and improved performance in comparison with the usually used ring and Rabenseifner algorithms.