Fast Energy-Optimal Multikernel DNN-Like Application Allocation on Multi-FPGA Platforms

Junnan Shan,Mihai T. Lazarescu,Jordi Cortadella,Luciano Lavagno,Mario R. Casu

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems（2022）

引用 1|浏览26

暂无评分

摘要

Platforms with multiple field-programmable gate arrays (FPGAs), such as Amazon Web Services (AWS) F1 instances, can efficiently accelerate multikernel pipelined applications, e.g., convolutional neural networks for machine vision tasks or transformer networks for natural language processing tasks. To reduce energy consumption when the FPGAs are underutilized, we propose a model to 1) find offline ...

查看译文

关键词

Field programmable gate arrays,Kernel,Copper,Resource management,Data transfer,Throughput,Data models

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要