Best Paper Award at MICRO’22
Rahul Bera, doctoral student in the SAFARI Research Group, won the Best Paper Award together with co-authors for their work “Hermes: Accelerating Long-Latency Load Requests via Perceptron-Based Off-Chip Load Prediction” at the MICRO’22.
The key idea behind Hermes is to: (1) accurately predict which load requests might go to off-chip, and (2) speculatively start fetching the data required by the predicted off-chip loads directly from the main memory in parallel to the cache accesses. Hermes proposes a lightweight, perceptron-based off-chip predictor that identifies off-chip load requests using multiple disparate programme features. The predictor is implemented using only tables and simple arithmetic operations like increment and decrement.
Paper on arXiv: external page https://arxiv.org/abs/2209.00188
Congratulations!