Algorithm-hardware co-design for Energy-Efficient A/D conversion in ReRAM-based accelerators
Chenguang Zhang, Zhihang Yuan, Xingchen Li, Guangyu Sun
备注:
6 pages; 7 figures; to be published in DATE 2024 - Design, Automation and Test in Europe; revised version: Add some necessary context and revise some equations into a clearer form
深度神经网络已被广泛部署于众多领域。凭借阻变随机存取存储器(Resistive Random Access Memory,ReRAM)交叉阵列的原位计算(即存内计算)能力,基于ReRAM的加速器在以低功耗、高性能加速深度神经网络(DNN)方面展现出潜力。然而,尽管具备功耗优势,这类加速器仍受限于外围电路的高功耗问题,尤其是模数转换器(Analog-to-Digital Converter,ADC),其功耗占总功耗的60%以上,这一问题阻碍了基于ReRAM的加速器实现更高的能效。\n 部分冗余的模数转换操作对维持推理精度没有贡献,可通过修改ADC搜索逻辑消除这类操作。基于这一观察,我们提出了一种算法-硬件协同设计方法,并从硬件设计与量化算法两方面探索协同设计路径。首先,我们聚焦于交叉阵列位线的输出分布,识别出细粒度的冗余ADC采样位。为进一步压缩ADC位数,我们提出了一种硬件友好的量化方法与编码方案,对不同区间的部分结果采用不同的量化策略。为支撑上述两项特性,我们提出了一种基于逐次逼近寄存器型模数转换器(SAR-ADC)的轻量级架构设计。值得一提的是,我们的方法不仅能效更高,还保留了算法的灵活性。实验表明,该方法可实现约1.6~2.3倍的ADC功耗降低。
英文摘要:
Deep neural networks are widely deployed in many fields. Due to the in-situ computation (known as processing in memory) capacity of the Resistive Random Access Memory (ReRAM) crossbar, ReRAM-based accelerator shows potential in accelerating DNN with low power and high performance. However, despite power advantage, such kind of accelerators suffer from the high power consumption of peripheral circuits, especially Analog-to-Digital Converter (ADC), which account for over 60 percent of total power consumption. This problem hinders the ReRAM-based accelerator to achieve higher efficiency.
Some redundant Analog-to-Digital conversion operations have no contribution to maintaining inference accuracy, and such operations can be eliminated by modifying the ADC searching logic. Based on such observations, we propose an algorithm-hardware co-design method and explore the co-design approach in both hardware design and quantization algorithms. Firstly, we focus on the distribution output along the crossbar's bit-lines and identify the fine-grained redundant ADC sampling bits. % of weight and To further compress ADC bits, we propose a hardware-friendly quantization method and coding scheme, in which different quantization strategy was applied to the partial results in different intervals. To support the two features above, we propose a lightweight architectural design based on SAR-ADC\@. It's worth mentioning that our method is not only more energy efficient but also retains the flexibility of the algorithm. Experiments demonstrate that our method can reduce about $1.6 \sim 2.3 \times$ ADC power reduction.
31.1 A 14.08-to-135.69Token/s ReRAM-on-Logic Stacked Outlier-Free Large-Language-Model Accelerator with Block-Clustered Weight-Compression and Adaptive Parallel-Speculative-Decoding
Hong Kong University of Science and Technology(香港科技大学);AI Chip Center for Emerging Smart System(新兴智能系统芯片中心);Hefei Reliance Memory(合肥芯联集成);Zhejiang University(浙江大学)
FORMS: Fine-grained Polarized ReRAM-based In-situ Computation for Mixed-signal DNN Accelerator
Northeastern University(东北大学);Georgia Institute of Technology(佐治亚理工学院);Samsung(三星);Stevens Institute of Technology(史蒂文斯理工学院);University of Southern California(南加州大学);University of Utah(犹他大学);University of Connecticut(康涅狄格大学)
Neural-PIM: Efficient Processing-In-Memory with Neural Approximation of Peripherals
Washington University in St. Louis(圣路易斯华盛顿大学);Shanghai Jiao Tong University(上海交通大学);Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所)
You Only Charge Once 2.0 : A End-to-End Analog Computing-in-Memory Architecture with Reconfigurable Switched Capacitors
The Hong Kong University of Science and Technology(香港科技大学);AI Chip Center for Emerging Smart Systems (ACCESS), InnoHK, Hong Kong(香港创新署新兴智能系统芯片中心);Huazhong University of Science and Technology(华中科技大学);Shenzhen University of Advanced Technology(深圳先进技术研究院)
IMAGINE: An 8-to-1b 22nm FD-SOI Compute-In-Memory CNN Accelerator With an End-to-End Analog Charge-Based 0.15-8POPS/W Macro Featuring Distribution-Aware Data Reshaping
Delft University of Technology(代尔夫特理工大学);UCLouvain(鲁汶大学)