3 papers
cs.AR2026
Enhancing Instruction Prefetching via Cache and TLB Management
Alexandre Valentin Jamet, Georgios Vavouliotis, Marti Torrents +2
Modern server workloads exhibit massive instruction footprints that heavily pressure the processor front-end, making L1 instruction (L1I) prefetching critical for sustaining perfor…
cs.MS2024
An Open-Source Framework for Efficient Numerically-Tailored Computations
Louis Ledoux, Marc Casas
We present a versatile open-source framework designed to facilitate efficient, numerically-tailored Matrix-Matrix Multiplications (MMMs). The framework offers two primary contribut…
cs.AR2024
A Two Level Neural Approach Combining Off-Chip Prediction with Adaptive Prefetch Filtering
Alexandre Valentin Jamet, Georgios Vavouliotis, Daniel A. Jiménez +2
To alleviate the performance and energy overheads of contemporary applications with large data footprints, we propose the Two Level Perceptron (TLP) predictor, a neural mechanism t…