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 2109.01404cs.AR
收藏

利用模拟存内计算进行端到端100- TOPS/W推理:我们到了吗?

End-to-end 100-TOPS/W Inference With Analog In-Memory Computing: Are We There Yet?

Gianmarco Ottavi, Geethan Karunaratne, Francesco Conti, Irem Boybat, Luca Benini, Davide Rossi

摘要:

内存加速(IMA)有望在深度神经网络(DNN)推理中实现重大效率提升,但IMA在数字系统中的集成仍面临挑战。我们提出了一个异构架构耦合8个RISC-V内核与IMA在共享内存集群,分析的好处和权衡存内计算的MobileNetV 2瓶颈层的实际用例。我们探讨了几种IMA集成策略,分析了性能,面积和能源效率。我们表明,虽然逐点层实现显着的速度比软件实现,dependency层无法有效地映射加速器上的参数导致吞吐量和面积之间的显着权衡。我们提出了一种混合解决方案,其中逐点卷积在IMA上执行,而依赖于集群核心,与具有类似性能的全集成IMA解决方案相比,SW执行速度提高了3倍,同时节省了50%的面积。

英文摘要:

In-Memory Acceleration (IMA) promises major efficiency improvements in deep neural network (DNN) inference, but challenges remain in the integration of IMA within a digital system. We propose a heterogeneous architecture coupling 8 RISC-V cores with an IMA in a shared-memory cluster, analyzing the benefits and trade-offs of in-memory computing on the realistic use case of a MobileNetV2 bottleneck layer. We explore several IMA integration strategies, analyzing performance, area, and energy efficiency. We show that while pointwise layers achieve significant speed-ups over software implementation, on depthwise layer the inability to efficiently map parameters on the accelerator leads to a significant trade-off between throughput and area. We propose a hybrid solution where pointwise convolutions are executed on IMA while depthwise on the cluster cores, achieving a speed-up of 3x over SW execution while saving 50% of area when compared to an all-in IMA solution with similar performance.

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。

↑