CeDMA: Enhancing Memory Efficiency of Heterogeneous Accelerator Systems Through Central DMA Controlling
- 4 November 2025
- book chapter
- Published by Springer Nature
- p. 129-144
Abstract
No abstract availableKeywords
This publication has 24 references indexed in Scilit:
- Exploration of TPU Architectures for the Optimized Transformer in Drainage Crossing Detection2024
- ReGraph: Scaling Graph Processing on HBM-enabled FPGAs with Heterogeneous Pipelines2022
- Accelerating Graph Convolutional Networks Using Crossbar-based Processing-In-Memory Architectures2022
- Ascend: a Scalable and Unified Architecture for Ubiquitous Deep Neural Network Computing : Industry Track Paper2021
- A Heterogeneous PIM Hardware-Software Co-Design for Energy-Efficient Graph Processing2020
- Shuhai: Benchmarking High Bandwidth Memory On FPGAS2020
- HyGCN: A GCN Accelerator with Hybrid Architecture2020
- AccPar: Tensor Partitioning for Heterogeneous Deep Learning Accelerators2020
- HBM (High Bandwidth Memory) DRAM Technology and Architecture2017
- Deep Residual Learning for Image Recognition2016