StreamDCIM：一种面向多模态Transformer的基于Tile的流式数字CIM加速器，采用混合驻留交叉转发数据流
StreamDCIM: A Tile-based Streaming Digital CIM Accelerator with Mixed-stationary Cross-forwarding Dataflow for Multimodal Transformer
AI总结：
本文提出StreamDCIM加速器，通过可重构CIM宏微架构、混合驻留交叉转发数据流和细粒度计算重写流水线，解决现有多模态Transformer加速器灵活性不足的问题，实现显著性能提升。
摘要：
多模态Transformer是新兴的人工智能（AI）模型，旨在处理来自多种模态的混合信号。数字存内计算（CIM）架构被认为在保持高精度的同时实现高效率方面具有前景。然而，当前基于数字CIM的加速器在微架构、数据流和流水线方面缺乏灵活性，难以有效加速多模态Transformer。本文提出StreamDCIM，一种面向多模态Transformer的基于Tile的流式数字CIM加速器。它通过三个特性克服上述挑战：第一，提出一种基于Tile的可重构CIM宏微架构，具有普通和混合可重构模式，以提高宏内CIM利用率。第二，实现一种混合驻留交叉转发数据流，并通过基于Tile的执行解耦来利用Tile级计算并行性。第三，引入一种类似乒乓的细粒度计算重写流水线，以重叠高延迟的片上CIM重写。实验结果表明，在典型多模态Transformer模型上，StreamDCIM相比非流式和基于层的流式CIM方案分别实现了2.63倍和1.28倍的几何平均性能提升。
英文摘要：
Multimodal Transformers are emerging artificial intelligence (AI) models designed to process a mixture of signals from diverse modalities. Digital computing-in-memory (CIM) architectures are considered promising for achieving high efficiency while maintaining high accuracy. However, current digital CIM-based accelerators exhibit inflexibility in microarchitecture, dataflow, and pipeline to effectively accelerate multimodal Transformer. In this paper, we propose StreamDCIM, a tile-based streaming digital CIM accelerator for multimodal Transformers. It overcomes the above challenges with three features: First, we present a tile-based reconfigurable CIM macro microarchitecture with normal and hybrid reconfigurable modes to improve intra-macro CIM utilization. Second, we implement a mixed-stationary cross-forwarding dataflow with tile-based execution decoupling to exploit tile-level computation parallelism. Third, we introduce a ping-pong-like fine-grained compute-rewriting pipeline to overlap high-latency on-chip CIM rewriting. Experimental results show that StreamDCIM outperforms non-streaming and layer-based streaming CIM-based solutions by geomean 2.63$\times$ and 1.28$\times$ on typical multimodal Transformer models.
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。
