paper-with-me

Papers

MassiveGNN: Efficient Training via Prefetching for Massively Connected Distributed Graphs

2024-10-30 · Aishwarya Sarkar, Sayan Ghosh, Nathan R. Tallent, Ali Jannesari

Graph Neural Networks (GNN) are indispensable in learning from graph-structured data, yet their rising computational costs, especially on massively connected graphs, pose significant challenges in terms of execution performance. To tackle this, distributed-memory solutions such as partitioning the graph to concurrently train multiple replicas of GNNs are in practice. However, approaches requiring a partitioned graph usually suffer from communication overhead and load imbalance, even under optimal partitioning and communication strategies due to irregularities in the neighborhood minibatch sampling. This paper proposes practical trade-offs for improving the sampling and communication overheads for representation learning on distributed graphs (using popular GraphSAGE architecture) by developing a parameterized continuous prefetch and eviction scheme on top of the state-of-the-art Amazon DistDGL distributed GNN framework, demonstrating about 15-40% improvement in end-to-end training performance on the National Energy Research Scientific Computing Center's (NERSC) Perlmutter supercomputer for various OGB datasets.

📄 PDF Abstract BibTeX arXiv:2410.22697

Code (1)

pnnl/MassiveGNN 공식 구현 pytorch

Tasks

Representation Learning

Methods 이 논문이 사용한 방법론

DistDGL DistDGL is a system for training GNNs in a mini-batch fashion on a cluster of machines. It is is based on the Deep Graph Library (DGL), a popular GNN development framework.…
GraphSAGE GraphSAGE is a general inductive framework that leverages node feature information (e.g., text attributes) to efficiently generate node embeddings for previously unseen…

Similar Papers 제목 키워드 기반

FedFetch: Faster Federated Learning with Adaptive Downstream Prefetching

2025-04-21 · Qifan Yan, Andrew Liu, Shiqi He, Mathias Lécuyer 외

Federated learning (FL) is a machine learning paradigm that facilitates massively distributed model training with end-user data on edge devices directed by a central server. However, the large number of heterogeneous cli…

Federated Learning

Rudder: Steering Prefetching in Distributed GNN Training using LLM Agents

2026-02-26 · Aishwarya Sarkar, Sayan Ghosh, Nathan Tallent, Aman Chadha 외 arxiv

Large-scale Graph Neural Networks (GNNs) are typically trained by sampling a vertex's neighbors to a fixed distance. Because large input graphs are distributed, training requires frequent irregular communication that sta…

RapidGNN: Energy and Communication-Efficient Distributed Training on Large-Scale Graph Neural Networks

2025-09-05 · Arefin Niam, Tevfik Kosar, M S Q Zulkar Nine arxiv

Graph Neural Networks (GNNs) have become popular across a diverse set of tasks in exploring structural relationships between entities. However, due to the highly connected structure of the datasets, distributed training …

Massively Distributed Antenna Systems with Non-Ideal Optical Fiber Front-hauls: A Promising Technology for 6G Wireless Communication Systems

2020-08-18 · Lisu Yu, Jingxian Wu, Andong Zhou, Erik G. Larsson 외

Employing massively distributed antennas brings radio access points (RAPs) closer to users, thus enables aggressive spectrum reuse that can bridge gaps between the scarce spectrum resource and extremely high connection d…

Distributed Learning in the Non-Convex World: From Batch to Streaming Data, and Beyond

2020-01-14 · Tsung-Hui Chang, Mingyi Hong, Hoi-To Wai, Xinwei Zhang 외

Distributed learning has become a critical enabler of the massively connected world envisioned by many. This article discusses four key elements of scalable distributed processing and real-time intelligence --- problems,…