论文标题
新兴NVM技术的统一表征平台：神经网络应用程序使用现成的NVM芯片进行基准测试
Unified Characterization Platform for Emerging NVM Technology: Neural Network Application Benchmarking Using off-the-shelf NVM Chips
论文作者
论文摘要
在本文中，我们提出了一个基于统一的基于FPGA的电测试台，用于表征不同的新出现的非易失性记忆（NVM）芯片。特别是，我们提出了多个市售，现成的NVM芯片的详细的电气表征和基准测试，即：MRAM，FERAM，CBRAM和RERAM。我们研究了重要的NVM参数，例如：（i）电流消耗模式，（ii）耐力和（iii）错误表征。然后将基于FPGA的测试台用于概念验证（POC）神经网络（NN）图像分类应用程序。在推理模式下，将四个新兴的NVM芯片用于AI应用程序作为主动重量存储器的标准SRAM和Flash Technology的基准测试。
In this paper, we present a unified FPGA based electrical test-bench for characterizing different emerging NonVolatile Memory (NVM) chips. In particular, we present detailed electrical characterization and benchmarking of multiple commercially available, off-the-shelf, NVM chips viz.: MRAM, FeRAM, CBRAM, and ReRAM. We investigate important NVM parameters such as: (i) current consumption patterns, (ii) endurance, and (iii) error characterization. The proposed FPGA based testbench is then utilized for a Proof-of-Concept (PoC) Neural Network (NN) image classification application. Four emerging NVM chips are benchmarked against standard SRAM and Flash technology for the AI application as active weight memory during inference mode.
