Bouquet of instruction pointers: Instruction pointer classifier-based spatial hardware prefetching

S Pakalapati, B Panda - 2020 ACM/IEEE 47th Annual …, 2020 - ieeexplore.ieee.org
Hardware prefetching is one of the common off-chip DRAM latency hiding techniques.
Though hardware prefetchers are ubiquitous in the commercial machines and prefetching …

Clip: Load criticality based data prefetching for bandwidth-constrained many-core systems

B Panda - Proceedings of the 56th Annual IEEE/ACM …, 2023 - dl.acm.org
Hardware prefetching is a latency-hiding technique that hides the costly off-chip DRAM
accesses. However, state-of-the-art prefetchers fail to deliver performance improvement in …

Micro-armed bandit: lightweight & reusable reinforcement learning for microarchitecture decision-making

G Gerogiannis, J Torrellas - Proceedings of the 56th Annual IEEE/ACM …, 2023 - dl.acm.org
Online Reinforcement Learning (RL) has been adopted as an effective mechanism in
various decision-making problems in microarchitecture. Its high adaptability and the ability to …

Berti: an accurate local-delta data prefetcher

A Navarro-Torres, B Panda… - 2022 55th IEEE/ACM …, 2022 - ieeexplore.ieee.org
Data prefetching is a technique that plays a crucial role in modern high-performance
processors by hiding long latency memory accesses. Several state-of-the-art hardware …

Managing prefetchers with deep reinforcement learning

M Jalili, M Erez - IEEE Computer Architecture Letters, 2022 - ieeexplore.ieee.org
We aim to reduce contention caused by multiple aggressive prefetchers on shared
resources (eg, LLC and memory bandwidth) with a multi-agent reinforcement learning …

Puppeteer: A random forest based manager for hardware prefetchers across the memory hierarchy

F Eris, M Louis, K Eris, J Abellan, A Joshi - ACM Transactions on …, 2022 - dl.acm.org
Over the years, processor throughput has steadily increased. However, the memory
throughput has not increased at the same rate, which has led to the memory wall problem in …

Lightweight ml-based runtime prefetcher selection on many-core platforms

ES Alcorta, M Madhav, S Tetrick, NJ Yadwadkar… - arXiv preprint arXiv …, 2023 - arxiv.org
Modern computer designs support composite prefetching, where multiple individual
prefetcher components are used to target different memory access patterns. However …

ZPP: a dynamic technique to eliminate cache pollution in NoC based MPSoCs

D Deb, J Jose - ACM Transactions on Embedded Computing Systems, 2023 - dl.acm.org
Data prefetching efficiently reduces the memory access latency in NUCA architectures as
the Last Level Cache (LLC) is shared and distributed across multiple cores. But cache …

A prefetch control strategy based on improved hill-climbing method in asymmetric multi-core architecture

J Fang, Y Xu, H Kong, M Cai - The Journal of Supercomputing, 2023 - Springer
Cache prefetching is a traditional way to reduce memory access latency. In multi-core
systems, aggressive prefetching may harm the system. In the past, prefetching throttling …

A prefetch-adaptive intelligent cache replacement policy based on machine learning

HJ Yang, J Fang, M Cai, Z Cai - Journal of Computer Science and …, 2023 - Springer
Hardware prefetching and replacement policies are two techniques to improve the
performance of the memory subsystem. While prefetching hides memory latency and …