@article{ART003090585},
author={Hyein Lee and Jinoo Joung},
title={Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2024},
volume={29},
number={6},
pages={1-12},
doi={10.9708/jksci.2024.29.06.001}
TY - JOUR
AU - Hyein Lee
AU - Jinoo Joung
TI - Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch
JO - Journal of The Korea Society of Computer and Information
PY - 2024
VL - 29
IS - 6
PB - The Korean Society Of Computer And Information
SP - 1
EP - 12
SN - 1598-849X
AB - Deep learning with faster and more accurate results requires large amounts of storage space and large computations. Accordingly, many studies are using hardware accelerators for quick and accurate calculations. However, the performance bottleneck is due to data movement between the hardware accelerators and the CPU. In this paper, we propose a data prefetch strategy that can efficiently reduce such operational bottlenecks. The core idea of the data prefetch strategy is to predict the data needed for the next task and upload it to local memory while the hardware accelerator (Matrix Multiplication Unit, MMU) performs a task. This strategy can be enhanced by using a dual buffer to perform read and write operations simultaneously. This reduces latency and execution time of data transfer. Through simulations, we demonstrate a 24% improvement in the performance of hardware accelerators by maximizing parallel processing with dual buffers and bottlenecks between memories with data prefetch.
KW - HW Accelerator;CNN;Bottleneck;Data Prefetch;Dual Buffer;RTL Simulation
DO - 10.9708/jksci.2024.29.06.001
ER -
Hyein Lee and Jinoo Joung. (2024). Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch. Journal of The Korea Society of Computer and Information, 29(6), 1-12.
Hyein Lee and Jinoo Joung. 2024, "Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch", Journal of The Korea Society of Computer and Information, vol.29, no.6 pp.1-12. Available from: doi:10.9708/jksci.2024.29.06.001
Hyein Lee, Jinoo Joung "Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch" Journal of The Korea Society of Computer and Information 29.6 pp.1-12 (2024) : 1.
Hyein Lee, Jinoo Joung. Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch. 2024; 29(6), 1-12. Available from: doi:10.9708/jksci.2024.29.06.001
Hyein Lee and Jinoo Joung. "Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch" Journal of The Korea Society of Computer and Information 29, no.6 (2024) : 1-12.doi: 10.9708/jksci.2024.29.06.001
Hyein Lee; Jinoo Joung. Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch. Journal of The Korea Society of Computer and Information, 29(6), 1-12. doi: 10.9708/jksci.2024.29.06.001
Hyein Lee; Jinoo Joung. Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch. Journal of The Korea Society of Computer and Information. 2024; 29(6) 1-12. doi: 10.9708/jksci.2024.29.06.001
Hyein Lee, Jinoo Joung. Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch. 2024; 29(6), 1-12. Available from: doi:10.9708/jksci.2024.29.06.001
Hyein Lee and Jinoo Joung. "Resolving Memory Bottlenecks in Hardware Accelerators with Data Prefetch" Journal of The Korea Society of Computer and Information 29, no.6 (2024) : 1-12.doi: 10.9708/jksci.2024.29.06.001
