Publication: Berti: An Accurate Local-Delta Data Prefetcher
Authors
Navarro-Torres, Agustín ; Panda, Biswabandan ; Alastruey-Benedé, Jesús ; Ibáñez, Pablo ; Viñals-Yúfera, Victor ; Ros, Alberto
item.page.secondaryauthor
item.page.director
Publisher
IEEE
publication.page.editor
publication.page.department
DOI
https://doi.org/10.1109/MICRO56248.2022.00072
item.page.type
info:eu-repo/semantics/article
Description
© 2022. IEEE.
This document is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0
This document is the accepted version of a published work that appeared in final form in 2022 55th IEEE/ACM International Symposium on Microarchitecture (MICRO).
To access the final work, see DOI: 10.1109/MICRO56248.2022.00072
Abstract
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 prefetchers exploit the concept of deltas, defined as the difference between the cache line addresses of two demand accesses. Existing delta prefetchers, such as best offset prefetching (BOP) and multi-lookahead prefetching (MLOP), train and predict future accesses based on global deltas. We observed that the use of global deltas results in missed opportunities to anticipate memory accesses.
In this paper, we propose Berti, a first-level data cache prefetcher that selects the best local deltas, i.e., those that consider only demand accesses issued by the same instruction. Thanks to a high-confidence mechanism that precisely detects the timely local deltas with high coverage, Berti generates accurate prefetch requests. Then, it orchestrates the prefetch requests to the memory hierarchy, using the selected deltas.
Our empirical results using ChampSim and SPEC CPU2017 and GAP workloads show that, with a storage overhead of just 2.55 KB, Berti improves performance by 8.5% compared to a baseline IP-stride and 3.5% compared to IPCP, a state-of-the-art prefetcher. Our evaluation also shows that Berti reduces dynamic energy at the memory hierarchy by 33.6% compared to IPCP, thanks to its high prefetch accuracy.
publication.page.subject
Citation
item.page.embargo
Collections
Ir a Estadísticas
Este ítem está sujeto a una licencia Creative Commons. http://creativecommons.org/licenses/by-nc-nd/4.0/