paper-with-me

홈 › Papers

OpenGeMM: A High-Utilization GeMM Accelerator Generator with Lightweight RISC-V Control and Tight Memory Coupling

2024-11-14 · Xiaoling Yi, Ryan Antonio, Joren Dumoulin, Jiacong Sun, Josse Van Delm, Guilherme Paim, Marian Verhelst

Deep neural networks (DNNs) face significant challenges when deployed on resource-constrained extreme edge devices due to their computational and data-intensive nature. While standalone accelerators tailored for specific application scenarios suffer from inflexible control and limited programmability, generic hardware acceleration platforms coupled with RISC-V CPUs can enable high reusability and flexibility, yet typically at the expense of system level efficiency and low utilization. To fill this gap, we propose OpenGeMM, an open-source acceleration platform, jointly demonstrating high efficiency and utilization, as well as ease of configurability and programmability. OpenGeMM encompasses a parameterized Chisel-coded GeMM accelerator, a lightweight RISC-V processor, and a tightly coupled multi-banked scratchpad memory. The GeMM core utilization and system efficiency are boosted through three mechanisms: configuration pre-loading, input pre-fetching with output buffering, and programmable strided memory access. Experimental results show that OpenGeMM can consistently achieve hardware utilization ranging from 81.89% to 99.34% across diverse CNN and Transformer workloads. Compared to the SotA open-source Gemmini accelerator, OpenGeMM demonstrates a 3.58x to 16.40x speedup on normalized throughput across a wide variety ofGeMM workloads, while achieving 4.68 TOPS/W system efficiency.

📄 PDF Abstract BibTeX arXiv:2411.09543

Code (1)

kuleuven-micas/snax_cluster 공식 구현

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Adam 설명 없음
Residual Connection 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

Gemmini: Enabling Systematic Deep-Learning Architecture Evaluation via Full-Stack Integration

2019-11-22 · Hasan Genc, Seah Kim, Alon Amid, Ameer Haj-Ali 외

DNN accelerators are often developed and evaluated in isolation without considering the cross-stack, system-level effects in real-world environments. This makes it difficult to appreciate the impact of System-on-Chip (So…

Deep Learning

WaferLLM: Large Language Model Inference at Wafer Scale

2025-02-06 · Congjie He, Yeqi Huang, Pei Mu, Ziming Miao 외

Emerging AI accelerators increasingly adopt wafer-scale manufacturing technologies, integrating hundreds of thousands of AI cores in a mesh architecture with large distributed on-chip memory (tens of GB in total) and ult…

GPULanguage ModelingLanguage ModellingLarge Language Model+1

Panacea: Novel DNN Accelerator using Accuracy-Preserving Asymmetric Quantization and Energy-Saving Bit-Slice Sparsity

2024-12-13 · Dongyun Kam, Myeongji Yun, Sunwoo Yoo, Seungwoo Hong 외

Low bit-precisions and their bit-slice sparsity have recently been studied to accelerate general matrix-multiplications (GEMM) during large-scale deep neural network (DNN) inferences. While the conventional symmetric qua…

Quantization

N3H-Core: Neuron-designed Neural Network Accelerator via FPGA-based Heterogeneous Computing Cores

2021-12-15 · Yu Gong, Zhihan Xu, Zhezhi He, Weifeng Zhang 외

Accelerating the neural network inference by FPGA has emerged as a popular option, since the reconfigurability and high performance computing capability of FPGA intrinsically satisfies the computation demand of the fast-…

Quantization

Evaluating Spatial Accelerator Architectures with Tiled Matrix-Matrix Multiplication

2021-06-19 · Gordon E. Moon, Hyoukjun Kwon, Geonhwa Jeong, Prasanth Chatarasi 외

There is a growing interest in custom spatial accelerators for machine learning applications. These accelerators employ a spatial array of processing elements (PEs) interacting via custom buffer hierarchies and networks-…

Scheduling