ATC 2022

Meta Info

2022 USENIX Annual Technical Conference

Homepage: https://www.usenix.org/conference/atc22

Paper list: https://www.usenix.org/conference/atc22/technical-sessions

Papers

DNN Inference

• DVABatch: Diversity-aware Multi-Entry Multi-Exit Batching for Efficient Processing of DNN Services on GPUs [Personal Note, Code]

• Serving Heterogeneous Machine Learning Models on Multi-GPU Servers with Spatio-Temporal Sharing [Personal Note, Code]

• SOTER: Guarding Black-box Inference for General Neural Networks at the Edge [Personal Note, Code]

DNN Training

• Whale: Efficient Giant Model Training over Heterogeneous GPUs [Personal Note, Code]

Resource Manager

• PilotFish: Harvesting Free Cycles of Cloud Gaming with Deep Learning Training [Personal Note]

• Memory Harvesting in Multi-GPU Systems with Hierarchical Unified Virtual Memory [Personal Note]