arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.01827cs.AR
收藏

在RISC-V向量架构中集成数字存内计算以加速深度学习

In-Pipeline Integration of Digital In-Memory-Computing into RISC-V Vector Architecture to Accelerate Deep Learning

Politecnico di Milano(米兰理工大学);STMicroelectronics(意法半导体)

Tommaso Spagnolo, Cristina Silvano, Riccardo Massa, Filippo Grillotti, Thomas Boesch, Giuseppe Desoli

AI总结:

本文提出在RISC-V向量架构中集成数字存内计算单元,通过专用指令提升深度学习推断效率,实现217倍加速和50倍面积归一化加速。

摘要:

扩展深度学习应用到边缘计算需求架构能够提供高计算性能和效率的同时遵守严格的功率和内存限制。数字存内计算(DIMC)通过将部分计算直接移动到内存数组中,显著减少数据移动并提高能效。本文介绍了一种新的架构,扩展了向量RISC-V指令集架构(ISA),以将紧密耦合的DIMC单元直接集成到流水线的执行阶段中,以加速边缘深度学习推断。具体而言,所提出的方法添加了四个专用指令,用于数据加载、计算和写回,使在目标架构上灵活和优化地控制推断执行成为可能。实验结果表明,在向量RISC-V中DIMC瓷砖的高利用率和在ResNet-50模型上的持续吞吐量,达到了137 GOP/s的峰值性能。所提出的架构在基准核心上实现了217倍的加速,并且在接近硬件资源极限时仍实现了50倍的面积归一化加速。实验结果证实了所提出架构作为可扩展且高效的解决方案,用于加速边缘深度学习推断的高潜力。

英文摘要:

Expanding Deep Learning applications toward edge computing demands architectures capable of delivering high computational performance and efficiency while adhering to tight power and memory constraints. Digital In-Memory Computing (DIMC) addresses this need by moving part of the computation directly within memory arrays, significantly reducing data movement and improving energy efficiency. This paper introduces a novel architecture that extends the Vector RISC-V Instruction Set Architecture (ISA) to integrate a tightly coupled DIMC unit directly into the execution stage of the pipeline, to accelerate Deep Learning inference at the edge. Specifically, the proposed approach adds four custom instructions dedicated to data loading, computation, and write-back, enabling flexible and optimal control of the inference execution on the target architecture. Experimental results demonstrate high utilization of the DIMC tile in Vector RISC-V and sustained throughput across the ResNet-50 model, achieving a peak performance of 137 GOP/s. The proposed architecture achieves a speedup of 217x over the baseline core and 50x area-normalized speedup even when operating near the hardware resource limits. The experimental results confirm the high potential of the proposed architecture as a scalable and efficient solution to accelerate Deep Learning inference on the edge.

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。

↑