Deep Neural Networks (DNNs) have achieved remarkable success across various intelligent tasks but encounter performance and energy challenges in inference execution due to data movement bottlenecks. We introduce DataMaestro, a versatile and efficient data streaming unit that brings the decoupled access/execute architecture to DNN dataflow accelerators to address this issue. DataMaestro supports flexible and programmable access patterns to accommodate diverse workload types and dataflows, incorporates fine-grained prefetch and addressing mode switching to mitigate bank conflicts, and enables customizable on-the-fly data manipulation to reduce memory footprints and access counts. We integrate five DataMaestros with a Tensor Core-like GeMM accelerator and a Quantization accelerator into a RISC-V host system for evaluation. The FPGA prototype and VLSI synthesis results demonstrate that DataMaestro helps the GeMM core achieve nearly 100% utilization, which is 1.05-21.39x better than state-of-the-art solutions, while minimizing area and energy consumption to merely 6.43% and 15.06% of the total system.
@article{arxiv.2504.14091,
title = {DataMaestro: A Versatile and Efficient Data Streaming Engine Bringing Decoupled Memory Access To Dataflow Accelerators},
author = {Xiaoling Yi and Yunhao Deng and Ryan Antonio and Fanchen Kong and Guilherme Paim and Marian Verhelst},
journal= {arXiv preprint arXiv:2504.14091},
year = {2025}
}