FastGNAS: Accelerating and Scaling Graph Neural Architecture Search on Multi-GPUs via Ring-Based Model Migration
Summary: FastGNAS combines data/architecture parallelism via ring-based model migration, enabling synchronization-free multi-GPU GNAS at scale. Its intermediate-result caching, probabilistic reuse, and predictive load balancing deliver up to 6.18× acceleration with competitive accuracy. (summarized by gpt-5.6-luna on Jul 26 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Zhen Song (Shandong University)
- 2. Hao Li (Northeastern University)
- 3. Tianyi Li (Aalborg University)
- 4. Yu Gu (Northeastern University)
- 5. Yushuai Li (Aalborg University)
- 6. Yanfeng Zhang (Northeastern University)
- 7. Christian S. Jensen (Aalborg University)
- 8. Lizhen Cui (Shandong University)
- 9. Ge Yu (Northeastern University)
BibTeX Citation
@inproceedings{song_sigmod26,
title = {{FastGNAS: Accelerating and Scaling Graph Neural Architecture Search on Multi-GPUs via Ring-Based Model Migration}},
author = {Song, Zhen and Li, Hao and Li, Tianyi and Gu, Yu and Li, Yushuai and Zhang, Yanfeng and Jensen, Christian S. and Cui, Lizhen and Yu, Ge},
series = {{SIGMOD} '26},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3802059},
url = {https://dl.acm.org/doi/10.1145/3802059},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 10 of 10 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next