Database Paper Browser

Back to papers

SIMPLE: Efficient Temporal Graph Neural Network Training at Scale with Dynamic Data Placement

Summary: SIMPLE targets the CPU-GPU data-loading bottleneck in large-scale temporal GNN training. Key idea: dynamic data placement with a small GPU buffer plus pipeline optimizations, cutting loading cost up to 96.8% and speeding training 1.8x–3.8x over TGL. (summarized by gpt-5.4-mini on May 24 2026)

Paper ID
6937
Venue
SIGMOD
Year
2024
Pagerank
4.8616315e-05
Overall Rank
7,014 | 51.21%
DOI
10.1145/3654977

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 9 of 9 citing papers.

Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 13 of 13 cited papers.

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

Previous Page 1 / 1 Next

Semantically Similar Papers