arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.04036cs.AR
收藏

PICNIC:面向LLM推理加速的硅光互连芯粒,具备计算网络与存内计算能力

PICNIC: Silicon Photonic Interconnected Chiplets with Computational Network and In-memory Computing for LLM Inference Acceleration

National University of Singapore(新加坡国立大学)

Yue Jiet Chong, Yimin Wang, Zhen Wu, Xuanyao Fong

AI总结:

本文提出PICNIC加速器,通过硅光互连、存内计算和芯粒聚类与功耗门控方案,显著提升LLM推理的加速比与能效。

摘要:

本文提出一种基于3D堆叠芯粒的大语言模型(LLM)推理加速器,由非易失性存内计算处理单元(PE)和PE间计算网络(IPCN)组成,并通过硅光互连有效解决通信瓶颈。开发了一种LLM映射方案,以优化硬件调度和工作负载映射。仿真结果表明,在采用芯粒聚类与功耗门控方案(CCPG)之前,其相比Nvidia A100实现了3.95倍的加速和30倍的能效提升。此外,通过实施CCPG,该系统在支持更大模型时获得了进一步的可扩展性和能效提升,在相近吞吐量下相比Nvidia H100实现了57倍的能效提升。

英文摘要:

This paper presents a 3D-stacked chiplets based large language model (LLM) inference accelerator, consisting of non-volatile in-memory-computing processing elements (PEs) and Inter-PE Computational Network (IPCN), interconnected via silicon photonic to effectively address the communication bottlenecks. A LLM mapping scheme was developed to optimize hardware scheduling and workload mapping. Simulation results show it achieves $3.95\times$ speedup and $30\times$ efficiency improvement over the Nvidia A100 before chiplet clustering and power gating scheme (CCPG). Additionally, the system achieves further scalability and efficiency improvement with the implementation of CCPG to accommodate larger models, attaining $57\times$ efficiency improvement over Nvidia H100 at similar throughput.

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。

↑