一种基于相变存储器的64核混合信号存内计算芯片,用于深度神经网络推理
A 64-core mixed-signal in-memory compute chip based on phase-change memory for deep neural network inference
AI总结:
本文提出一种基于相变存储器的64核混合信号存内计算芯片,在14nm CMOS工艺上集成,实现ResNet和LSTM网络接近软件精度的推理,吞吐量达63.1 TOPS,能效9.76 TOPS/W。
摘要:
在人工神经网络的硬件实现中,反复将突触权重值从存储器搬运到处理单元一直是能源效率低下的关键来源。采用空间实例化突触权重的模拟存内计算(AIMC)通过在芯片上存储的网络权重内直接执行矩阵-向量乘法(MVM)来完成推理工作负载,有望克服这一挑战。然而,为了实现延迟和能耗的端到端改进,AIMC必须与片上数字运算和通信相结合,以迈向完全在芯片上实现整个推理工作负载的配置。此外,非常期望在不针对应用重新调整芯片的情况下实现高MVM和推理精度。在此,我们展示了一款多核AIMC芯片,该芯片采用14纳米互补金属氧化物半导体(CMOS)技术设计并制造,并集成了后端相变存储器(PCM)。这款完全集成的芯片具有64个256x256 AIMC核心,通过片上通信网络互连。它还实现了ResNet卷积神经网络和长短期记忆(LSTM)网络所涉及的数字激活函数和处理。我们展示了在芯片上实现与权重层和激活函数相关的所有计算的同时,ResNet和LSTM网络达到了接近软件等效的推理精度。该芯片在8位输入/输出矩阵-向量乘法下,可实现63.1 TOPS的最大吞吐量,能效为9.76 TOPS/W。
英文摘要:
The need to repeatedly shuttle around synaptic weight values from memory to processing units has been a key source of energy inefficiency associated with hardware implementation of artificial neural networks. Analog in-memory computing (AIMC) with spatially instantiated synaptic weights holds high promise to overcome this challenge, by performing matrix-vector multiplications (MVMs) directly within the network weights stored on a chip to execute an inference workload. However, to achieve end-to-end improvements in latency and energy consumption, AIMC must be combined with on-chip digital operations and communication to move towards configurations in which a full inference workload is realized entirely on-chip. Moreover, it is highly desirable to achieve high MVM and inference accuracy without application-wise re-tuning of the chip. Here, we present a multi-core AIMC chip designed and fabricated in 14-nm complementary metal-oxide-semiconductor (CMOS) technology with backend-integrated phase-change memory (PCM). The fully-integrated chip features 64 256x256 AIMC cores interconnected via an on-chip communication network. It also implements the digital activation functions and processing involved in ResNet convolutional neural networks and long short-term memory (LSTM) networks. We demonstrate near software-equivalent inference accuracy with ResNet and LSTM networks while implementing all the computations associated with the weight layers and the activation functions on-chip. The chip can achieve a maximal throughput of 63.1 TOPS at an energy efficiency of 9.76 TOPS/W for 8-bit input/output matrix-vector multiplications.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。