返回
Dynamic Precision-Scalable Thermal Mapping Algorithm for Three Dimensional Systolic-Array Based Neural Network Accelerator
DOI:10.1109/TCE.2024.3378706.png)
摘要
En 中文
Nowadays, the systolic-array based accelerator has been used widely for the neural-network applications. Multiple systolic-array based accelerator chips can be stacked by the 3D IC technology to improve the performance of the neural-network applications. However, the 3D accelerator increases the power density and causes the overheating. To avoid the overheating, the sacrifice of the performance for the 3D accelerator under the thermal limitations is important. In this work, a dynamic precision-scalable thermal mapping algorithm (DPSTM) is proposed to change the active processing elements with different data precisions in the 3D accelerators dynamically. The goal is to control the power density and peak temperature of the 3D accelerator. Compared with the related works, DPSTM can reduce 29%-77% and 7%-73% latencies in AlexNet and ResNet-18 with 92-95 degrees C thermal limitations.
Keyword:
Three-dimensional displays
Power system measurements
Density measurement
Artificial neural networks
Temperature distribution
Stacking
Heuristic algorithms
Neural network
accelerator
thermal mapping
3D
期刊
IF:
10.9
论文数:
5.3K
被引数:
6.8K
机构
引用论文
Rubber Compounds Based on Blends of Propylene Oxide Rubber and Ultrafine Polytetrafluoroethylene基于环氧丙烷橡胶与超细聚四氟乙烯共混物的橡胶复合材料
Pd-catalyzed C-N coupling of primary (het)arylamines with 5-substituted 3-chloro-4 H -1,2,6-thiadiazin-4-ones钯催化的一级(杂)芳胺与5-取代-3-氯-4H-1,2,6-噻二嗪-4-酮的C-N偶联
Temperature-Prediction Based Rate-Adjusted Time and Space Mapping Algorithm for 3D CNN Accelerator Systems基于温度预测的三维CNN加速器系统速率调整时空映射算法
Crystal structure of aminoglycoside 4'-O-adenylyltransferase ANT(4')-IIb, apo氨基糖苷类4'-O-腺苷酰转移酶ANT(4')-IIb的晶体结构,去质子化状态

