NMExplorer: An Efficient Exploration Framework for DIMM-based Near-Memory Tensor Reduction

Various DIMM-based near-memory processing (DIMM-NMP) architectures have been proposed to accelerate tensor reduction. With careful evaluation, we find that diverse scenarios exhibit distinct performance on DIMM-NMP architectures adopting different design configurations. However, given a tensor reduction scenario, there is a lack of a fast and accurate solution to identify a proper DIMM-NMP architecture. To tackle this problem, we propose an efficient exploration framework called NMExplorer. Given a scenario and hardware parameters, NMExplorer can generate and explore a wide range of potential design configurations. Experiments show that the recommended designs can outperform state-of-the-art DIMM-NMP accelerators by up to 1.95× in performance and 3.69× in energy.

Download paper here

@inproceedings{li2023nmexplorer,
  title={NMExplorer: An Efficient Exploration Framework for DIMM-based Near-Memory Tensor Reduction},
  author={Li, Cong and Zhou, Zhe and Li, Xingchen and Sun, Guangyu and Niu, Dimin},
  booktitle={2023 60th ACM/IEEE Design Automation Conference (DAC)},
  pages={1--6},
  year={2023},
  organization={IEEE}
}