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 …
various decision-making problems in microarchitecture. Its high adaptability and the ability to …
Start: Scalable tracking for any rowhammer threshold
The Rowhammer vulnerability is worsening, with the Rowhammer Threshold (T RH)
reducing from 139K to 4.8 K activations over the last decade. As thresholds reduce further …
reducing from 139K to 4.8 K activations over the last decade. As thresholds reduce further …
ImPress: Securing DRAM Against Data-Disturbance Errors via Implicit Row-Press Mitigation
DRAM cells are susceptible to Data-Disturbance Errors (DDE), which can be exploited by an
attacker to compromise system security. Rowhammer is a well-known DDE vulnerability that …
attacker to compromise system security. Rowhammer is a well-known DDE vulnerability that …
ReCon: Efficient Detection, Management, and Use of Non-Speculative Information Leakage
P Aimoniotis, AB Kvalsvik, X Chen… - Proceedings of the 56th …, 2023 - dl.acm.org
In a speculative side-channel attack, a secret is improperly accessed and then leaked by
passing it to a transmitter instruction. Several proposed defenses effectively close this …
passing it to a transmitter instruction. Several proposed defenses effectively close this …
A Two Level Neural Approach Combining Off-Chip Prediction with Adaptive Prefetch Filtering
AV Jamet, G Vavouliotis, DA Jiménez… - … Symposium on High …, 2024 - ieeexplore.ieee.org
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 …
data footprints, we propose the Two Level Perceptron (TLP) predictor, a neural mechanism …
Branch Target Buffer Organizations
To accommodate very large instruction footprints, modern high-performance processors rely
on fetch directed instruction prefetching through huge branch predictors and a hierarchy of …
on fetch directed instruction prefetching through huge branch predictors and a hierarchy of …
CHROME: Concurrency-aware holistic cache management framework with online reinforcement learning
Cache management is a critical aspect of computer architecture, encompassing techniques
such as cache replacement, bypassing, and prefetching. Existing research has often …
such as cache replacement, bypassing, and prefetching. Existing research has often …
Virtuoso: An open-source, comprehensive and modular simulation framework for virtual memory research
K Kanellopoulos, K Sgouras, O Mutlu - arXiv preprint arXiv:2403.04635, 2024 - arxiv.org
Virtual memory is a cornerstone of modern computing systems. Introduced as one of the
earliest instances of hardware-software co-design, VM facilitates programmer-transparent …
earliest instances of hardware-software co-design, VM facilitates programmer-transparent …
Attention, Distillation, and Tabularization: Towards Practical Neural Network-Based Prefetching
Attention-based Neural Networks (NN) have demonstrated their effectiveness in accurate
memory access prediction, an essential step in data prefetching. However, the substantial …
memory access prediction, an essential step in data prefetching. However, the substantial …
A New Formulation of Neural Data Prefetching
Temporal data prefetchers have the potential to produce significant performance gains by
prefetching irregular data streams. Recent work has introduced a neural model for temporal …
prefetching irregular data streams. Recent work has introduced a neural model for temporal …