ICP is a prefetcher that learns stable instruction correlations to speculatively compute future irregular memory accesses, outperforming Triangel by 14% and DMP by 6% with only 2.1 KB storage.
The gem5 simulator,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
Comprehensive profiling of expert selection in frontier MoE models reveals temporal and spatial patterns that enable 6.6x speedup on wafer-scale GPUs and 1.25x on existing systems via targeted optimizations.
citing papers explorer
-
ICP: Exploiting Instruction Correlation for Prefetching Irregular Memory Accesses
ICP is a prefetcher that learns stable instruction correlations to speculatively compute future irregular memory accesses, outperforming Triangel by 14% and DMP by 6% with only 2.1 KB storage.
-
Patterns behind Chaos: Forecasting Data Movement for Efficient Large-Scale MoE LLM Inference
Comprehensive profiling of expert selection in frontier MoE models reveals temporal and spatial patterns that enable 6.6x speedup on wafer-scale GPUs and 1.25x on existing systems via targeted optimizations.