AI 中文总结
针对2.5D PIM小芯片架构的电压降问题，提出基于LSTM-PDN代理的ReVolt框架，通过动态调整OU大小缓解电压降，EDP平均降低76倍且不影响推理精度。
AI 中文摘要
基于内存处理（PIM）的2.5D多小芯片平台是机器学习（ML）工作负载的支撑技术，但其性能受电源分配网络（PDN）影响：小芯片级电流需求的变化会引发空间和时间上变化的电压降，这类降事件会导致电压违例、降低系统性能，并影响ML工作负载的推理精度。本研究提出ReVolt，一种基于动态操作单元（OU）的框架，用于缓解基于PIM的多小芯片系统中的电压降。ReVolt利用基于LSTM的PDN代理在运行时预测每个小芯片的供电电压轨迹，支持主动调整OU大小以缓解降事件。通过将OU大小作为控制旋钮，ReVolt在维持计算精度的同时调节小芯片级电流需求，可防止电压降违例、改善能量延迟乘积（EDP）并保留ML模型的推理精度。实验结果表明，ReVolt在防止电压降违例的同时，与现有的固定和动态OU基准相比，EDP平均降低76倍，且不损害ML模型的推理精度。
英文摘要
Processing-in-memory (PIM)-based 2.5D multi-chiplet platforms are enablers for machine learning (ML) workloads. However, their performance is affected by the power delivery network (PDN), where varying chiplet-level current demand induces spatially and temporally varying voltage droop. These droop events lead to voltage violations, degrades system performance, and impact inference accuracy for ML workloads. In this work, we propose ReVolt, a dynamic operation unit (OU)-based framework for mitigating voltage droop in PIM-based multi-chiplet systems. ReVolt leverages an LSTM-based PDN surrogate to predict per-chiplet supply voltage trajectories at runtime, enabling proactive adjustment of OU size to mitigate droop events. By treating OU size as a control knob, ReVolt regulates chiplet-level current demand while maintaining computational accuracy. This approach prevents voltage droop violations and improves energy-delay product (EDP) while preserving ML model inference accuracy. Experimental results demonstrate that ReVolt prevents voltage droop violations while achieving an average 76x reduction in EDP compared to existing fixed and dynamic OU-based baselines, without compromising inference accuracy of ML models.
CommentsAccepted for Publication in IEEE/ACM Embedded Systems Week (ESWEEK-26)
