Public cloud providers seek to meet stringent performance requirements and low hardware cost. A key driver of performance and cost is main memory. Memory pooling promises to …
Past research has proposed numerous hardware prefetching techniques, most of which rely on exploiting one specific type of program context information (eg, program counter …
We present Decoupled Vector Runahead (DVR), an in-core prefetching technique, executing separately to the main application thread, that exploits massive amounts of …
Prefetching which predicts future memory accesses and preloads them from main memory, is a widely-adopted technique to overcome the processor-memory performance gap …
Machine learning algorithms have shown potential to improve prefetching performance by accurately predicting future memory accesses. Existing approaches are based on the …
M Kwon, S Lee, M Jung - Proceedings of the 15th ACM Workshop on …, 2023 - dl.acm.org
Integrating compute express link (CXL) with SSDs allows scalable access to large memory but has slower speeds than DRAMs. We present ExPAND, an expander-driven CXL …
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 …
H Li, K Liu, T Liang, Z Li, T Lu, H Yuan… - … Symposium on High …, 2023 - ieeexplore.ieee.org
Memory disaggregation is a promising direction to mitigate memory contention in datacenters. To make memory disaggregation practical, prior efforts expose remote memory …
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 …