paper-with-me

Papers

Accelerating AllReduce with a Persistent Straggler

2025-05-29 · Arjun Devraj, Eric Ding, Abhishek Vijaya Kumar, Robert Kleinberg, Rachee Singh

Distributed machine learning workloads use data and tensor parallelism for training and inference, both of which rely on the AllReduce collective to synchronize gradients or activations. However, bulk-synchronous AllReduce algorithms can be delayed by a persistent straggler that is slower to reach the synchronization barrier required to begin the collective. To address this challenge, we propose StragglAR: an AllReduce algorithm that accelerates distributed training and inference in the presence of persistent stragglers. StragglAR implements a ReduceScatter among the remaining GPUs during the straggler-induced delay, and then executes a novel collective algorithm to complete the AllReduce once the straggler reaches the synchronization barrier. StragglAR achieves a 2x theoretical speedup over popular bandwidth-efficient AllReduce algorithms (e.g., Ring) for large GPU clusters with persistent stragglers. On an 8-GPU server, our implementation of StragglAR yields a 22% speedup over state-of-the-art AllReduce algorithms.

📄 PDF Abstract BibTeX arXiv:2505.23523

Code (1)

arjundevraj/stragglar 공식 구현 pytorch

Tasks

GPU

Similar Papers 제목 키워드 기반

Don't Let a Few Network Failures Slow the Entire AllReduce

2026-06-01 · Peiqing Chen, Jiedong Jiang, Nengneng Yu, Yuefeng Wang 외 arxiv

Network failures are among the most frequent hardware faults in large-scale GPU clusters and a leading cause of training-job interruptions. Modern collective communication libraries such as NCCL mitigate network failures…

AntDT: A Self-Adaptive Distributed Training Framework for Leader and Straggler Nodes

2024-04-15 · Youshao Xiao, Lin Ju, Zhenglei Zhou, Siyuan Li 외

Many distributed training techniques like Parameter Server and AllReduce have been proposed to take advantage of the increasingly large data and rich features. However, stragglers frequently occur in distributed training…

Caramel: Accelerating Decentralized Distributed Deep Learning with Computation Scheduling

2020-04-29 · Sayed Hadi Hashemi, Sangeetha Abdu Jyothi, Brighten Godfrey, Roy Campbell

The method of choice for parameter aggregation in Deep Neural Network (DNN) training, a network-intensive task, is shifting from the Parameter Server model to decentralized aggregation schemes (AllReduce) inspired by the…

Deep LearningScheduling

CodedReduce: A Fast and Robust Framework for Gradient Aggregation in Distributed Learning

2019-02-06 · Amirhossein Reisizadeh, Saurav Prakash, Ramtin Pedarsani, Amir Salman Avestimehr

We focus on the commonly used synchronous Gradient Descent paradigm for large-scale distributed learning, for which there has been a growing interest to develop efficient and robust gradient aggregation strategies that o…

Stochastic Gradient Push for Distributed Deep Learning

2018-11-27 · ICLR 2019 5 · Mahmoud Assran, Nicolas Loizou, Nicolas Ballas, Michael Rabbat

Distributed data-parallel algorithms aim to accelerate the training of deep neural networks by parallelizing the computation of large mini-batch gradient updates across multiple nodes. Approaches that synchronize nodes u…

Deep LearningGeneral Classificationimage-classificationImage Classification+2