EMOGI: Efficient Memory-access for Out-of-memory Graph-traversal in GPUs
Summary: EMOGI enables out-of-memory graph traversal on GPUs by streaming directly from host memory at cache-line granularity, bypassing UVM. Coalescing external requests reduces PCIe traffic, enabling near full bandwidth and ~2.6x UVM speedup; scales with PCIe 4.0. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Seung Won Min (University of Illinois Urbana-Champaign)
- 2. Vikram Sharma Mailthody (University of Illinois Urbana-Champaign)
- 3. Zaid Qureshi (University of Illinois Urbana-Champaign)
- 4. Jinjun Xiong (IBM T.J. Watson Research Center)
- 5. Eiman Ebrahimi (NVIDIA)
- 6. Wen-mei Hwu (University of Illinois Urbana-Champaign)
BibTeX Citation
@article{min_vldb21,
title = {{EMOGI: Efficient Memory-access for Out-of-memory Graph-traversal in GPUs}},
author = {Min, Seung Won and Mailthody, Vikram Sharma and Qureshi, Zaid and Xiong, Jinjun and Ebrahimi, Eiman and Hwu, Wen-mei},
journal = {PVLDB},
series = {{VLDB} '21},
volume = {14},
number = {2},
pages = {114--127},
doi = {10.14778/3425879.3425883},
url = {https://doi.org/10.14778/3425879.3425883},
year = {2021}
}Incoming Citations (Sorted by Pagerank)
Showing 6 of 6 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,224 | Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture | 2021 | VLDB | 0.000114497 |
| 3,626 | Triton Join: Efficiently Scaling to a Large Join State on GPUs with Fast Interconnects | 2022 | SIGMOD | 7.1524537e-05 |
| 6,249 | Self-adaptive Graph Traversal on GPUs | 2021 | SIGMOD | 5.8358127e-05 |
| 6,924 | HongTu: Scalable Full-Graph GNN Training on Multiple GPUs | 2023 | SIGMOD | 5.6426299e-05 |
| 8,400 | TOD: GPU-accelerated Outlier Detection via Tensor Operations | 2023 | VLDB | 5.3395906e-05 |
| 11,338 | Efficient Graph Data Access for Out-of-Memory GPU Streaming Graph Processing | 2025 | VLDB | 4.9793485e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 6 of 6 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 239 | The Ubiquity of Large Graphs and Surprising Challenges of Graph Processing | 2018 | VLDB | 0.000235107 |
| 496 | From "Think Like a Vertex" to "Think Like a Graph" | 2014 | VLDB | 0.00017323054 |
| 1,997 | GraphMat: High performance graph analytics made productive | 2015 | VLDB | 9.2158333e-05 |
| 2,068 | Traversing Large Graphs on GPUs with Unified Memory | 2020 | VLDB | 9.0920811e-05 |
| 4,207 | GPU-based Graph Traversal on Compressed Graphs | 2019 | SIGMOD | 6.7343908e-05 |
| 4,237 | Accelerating Dynamic Graph Analytics on GPUs | 2018 | VLDB | 6.7107028e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 4,237 | Accelerating Dynamic Graph Analytics on GPUs | 2018 | VLDB |
| 2 | 9,957 | Towards Sufficient GPU-accelerated Dynamic Graph Management: Survey and Experiment | 2025 | VLDB |
| 3 | 11,338 | Efficient Graph Data Access for Out-of-Memory GPU Streaming Graph Processing | 2025 | VLDB |
| 4 | 5,172 | CGgraph: An Ultra-fast Graph Processing System on Modern Commodity CPU-GPU Co-processor | 2024 | VLDB |
| 5 | 2,468 | GPU-Accelerated Subgraph Enumeration on Partitioned Graphs | 2020 | SIGMOD |
| 6 | 2,068 | Traversing Large Graphs on GPUs with Unified Memory | 2020 | VLDB |
| 7 | 3,011 | Efficient GPU-Accelerated Subgraph Matching | 2023 | SIGMOD |
| 8 | 1,224 | Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture | 2021 | VLDB |
| 9 | 6,249 | Self-adaptive Graph Traversal on GPUs | 2021 | SIGMOD |
| 10 | 4,207 | GPU-based Graph Traversal on Compressed Graphs | 2019 | SIGMOD |