Scholarlink

Large-scale graph neural network (GNN) training systems on GPUs with CPU memory and storage face the challenge of efficiently caching the embedding data with accuracy guarantee. In this paper, we propose HCGNN, an out-of-GPU-memory GNN training system that combines GPU sampling and historical embedding caching. Our system supports dynamic embedding data caching through heuristic-based historical two-level cache design with lightweight data proactive eviction and high cache hit ratio. Compared with SOTA frameworks, HCGNN shows up to 6.7x speedup on graph sampling and 4.3x speedup on feature gathering within 0.5% accuracy loss.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Accelerating Large-Scale Out-of-GPU-Core GNN Training with Two-Level Historical Caching

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Jing Wang,
  • Taolei Wang,
  • Juntao Huang,
  • Yibo Liu,
  • Xinkai Wang,
  • Marius Kreutzer,
  • Chao Li,
  • Minyi Guo

摘要

Large-scale graph neural network (GNN) training systems on GPUs with CPU memory and storage face the challenge of efficiently caching the embedding data with accuracy guarantee. In this paper, we propose HCGNN, an out-of-GPU-memory GNN training system that combines GPU sampling and historical embedding caching. Our system supports dynamic embedding data caching through heuristic-based historical two-level cache design with lightweight data proactive eviction and high cache hit ratio. Compared with SOTA frameworks, HCGNN shows up to 6.7x speedup on graph sampling and 4.3x speedup on feature gathering within 0.5% accuracy loss.