eMamba: Efficient Acceleration Framework for Mamba Models in Edge Computing

Kim Jiyong

doi:10.1145/3762190

Return

eMamba: Efficient Acceleration Framework for Mamba Models in Edge Computing

2025-09-01

DOI：10.1145/3762190

Save

View PDF

Abstract

En 中文

State Space Model (SSM)-based machine learning architectures have recently gained significant attention for processing sequential data. Mamba, a recent sequence-to-sequence SSM, offers competitive accuracy with superior computational efficiency compared to state-of-the-art transformer models. While this advantage makes Mamba particularly promising for resource-constrained edge devices, no hardware acceleration frameworks are currently optimized for deploying it in such environments. This article presents eMamba, a comprehensive end-to-end hardware acceleration framework explicitly designed for deploying Mamba models on edge platforms. eMamba maximizes computational efficiency by replacing complex normalization layers with lightweight hardware-aware alternatives and approximating expensive operations, such as SiLU activation and exponentiation, considering the target applications. Then, it performs an approximation-aware neural architecture search (NAS) to tune the learnable parameters used during approximation. Evaluations with Fashion-MNIST, CIFAR-10, and MARS, an open-source human pose estimation dataset, show eMamba achieves comparable accuracy to state-of-the-art techniques using 1.63-19.9x fewer parameters. In addition, it generalizes well to large-scale natural language tasks, demonstrating stable perplexity across varying sequence lengths on the WikiText2 dataset. We also quantize and implement the entire eMamba pipeline on an AMD ZCU102 FPGA and ASIC using GlobalFoundries (GF) 22 nm technology. Experimental results show 4.95-5.62x lower latency and 2.22-9.95x higher throughput, with 4.77x smaller area, 9.84x lower power, and 48.6x lower energy consumption than baseline solutions while maintaining competitive accuracy.

Keywords:

Mamba

HW acceleration

edge computing

approximation

quantization

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.