HongTu: Scalable Full-Graph GNN Training on Multiple GPUs

Summary: HongTu scales full-graph GNN training on multi-GPU platforms with CPU-memory vertex storage, GPU offload, and recomputation caching. It minimizes host-GPU traffic with deduplicated communication and cost-guided reorganization, and delivers speedups over DistGNN. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: 6749
Venue: SIGMOD
Year: 2023
Pagerank: 4.8324242e-05
Overall Rank: 7,087 | 50.75%
DOI: 10.1145/3626733

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 8 of 8 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
7,609	Systems for Scalable Graph Analytics and Machine Learning: Trends and Methods	2025	VLDB	4.6921981e-05
10,027	NeutronHeter: Optimizing Distributed Graph Neural Network Training for Heterogeneous Clusters	2026	SIGMOD	4.1905499e-05
10,066	DepCache: A KV Cache Management Framework for GraphRAG with Dependency Attention	2026	SIGMOD	4.1905499e-05
10,310	NeutronCloud: Resource-Aware Distributed GNN Training in Fluctuating Cloud Environments	2026	VLDB	4.1905499e-05
10,548	Graph Neural Network Training Systems: A Performance Comparison of Full-Graph and Mini-Batch.	2025	VLDB	4.1905499e-05
10,579	NeutronTask: Scalable and Efficient Multi-GPU GNN Training with Task Parallelism	2025	VLDB	4.1905499e-05
10,742	Faster Convergence in Mini-batch Graph Neural Networks Training with Pseudo Full Neighborhood Compensation	2025	VLDB	4.1905499e-05
11,029	Improving Graph Compression for Efficient Resource-Constrained Graph Analytics	2024	VLDB	4.1905499e-05

Outgoing Citations (Sorted by Pagerank)

Showing 8 of 8 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
271	AliGraph: A Comprehensive Graph Neural Network Platform	2019	VLDB	0.00029565193
1,102	Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture	2021	VLDB	0.00014011556
1,162	Sancus: Staleness-Aware Communication-Avoiding Full-Graph Decentralized Training in Large-Scale Graph Neural Networks	2022	VLDB	0.00013573136
2,072	HippogriffDB: Balancing I/O and GPU Bandwidth in Big Data Analytics	2016	VLDB	9.6300019e-05
2,292	Pipelined Query Processing in Coprocessor Environments	2018	SIGMOD	9.0884645e-05
3,028	NeutronStar: Distributed GNN Training with Hybrid Dependency Management	2022	SIGMOD	7.6833093e-05
4,354	LargeEA: Aligning Entities for Large-scale Knowledge Graphs	2022	VLDB	6.2534656e-05
5,711	EMOGI: Efficient Memory-access for Out-of-memory Graph-traversal in GPUs	2021	VLDB	5.3603407e-05

Semantically Similar Papers

Overall Rank	Paper	Year	Venue	Pagerank
6,944	Efficient Training of Graph Neural Networks on Large Graphs	2024	VLDB	4.8875946e-05
9,400	NeutronTP: Load-Balanced Distributed Full-Graph GNN Training with Tensor Parallelism	2025	VLDB	4.3399748e-05
2,399	ByteGNN: Efficient Graph Neural Network Training at Large Scale	2022	VLDB	8.8869693e-05
3,988	G3: When Graph Neural Networks Meet Parallel Graph Processing Systems on GPUs	2020	VLDB	6.5548465e-05
5,746	Comprehensive Evaluation of GNN Training Systems: A Data Management Perspective	2024	VLDB	5.3429324e-05
10,579	NeutronTask: Scalable and Efficient Multi-GPU GNN Training with Task Parallelism	2025	VLDB	4.1905499e-05
1,102	Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture	2021	VLDB	0.00014011556
7,543	XGNN: Boosting Multi-GPU GNN Training via Global GNN Memory Store	2024	VLDB	4.7103673e-05
9,596	Scalable Graph Convolutional Network Training on Distributed-Memory Systems	2023	VLDB	4.3150788e-05
3,092	Scalable and Efficient Full-Graph GNN Training for Large Graphs	2023	SIGMOD	7.5869574e-05