Libra: Unleashing GPU Heterogeneity for High-Performance Sparse Matrix Multiplication

Documenti analoghi

• FlashSparse: Minimizing Computation Redundancy for Fast Sparse Matrix Multiplications on Tensor Cores
  di: Shi, Jinliang, et al.
  Pubblicazione: (2024)
• SparkAttention: High-Performance Multi-Head Attention for Large Models on Volta GPU Architecture
  di: Xu, Youxuan, et al.
  Pubblicazione: (2025)
• Near-Optimal Sparse Allreduce for Distributed Deep Learning
  di: Li, Shigang, et al.
  Pubblicazione: (2022)
• AutoDDL: Automatic Distributed Deep Learning with Near-Optimal Bandwidth Cost
  di: Chen, Jinfan, et al.
  Pubblicazione: (2023)
• Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines
  di: Li, Shigang, et al.
  Pubblicazione: (2021)