TTNNM: Thermal- and Traffic-Aware Neural Network Mapping on 3D-NoC-based Accelerator
Open Access
- 12 June 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 364-369
Abstract
3D Network on Chips (3D-NoCs) have ample on-chip wiring resources and high bandwidth, yet face numerous hotspots and higher temperature gradients due to increased integration and power density. This could lead to device failure, impacting system stability. Our paper introduces a thermal- and traffic-aware mapping method for 3D-NoC-based neural network accelerators. Firstly, based on the average load of different neural network layer, we determine their mapping sequences and suitable dies. Secondly, to minimize delay and alleviate hotspot temperatures, we allocate groups to appropriate nodes. Compared with previous works, TTNNM reduces the average temperature by 3.0°C, 2.2°C, 2.4°C, temperature variance by 58.4%, 64.8%, 73.0%, maximum temperature by 9.3°C, 7.9°C, 12.0°C, and packet latency by 31.7%, 17.2%, 25.1%.Keywords
Funding Information
- National Nature Science Foundation of China (62104098)
- Natural Science Foundation of Jiangsu Province for Youth (BK20210178)
- Joint Funds of the National Nature Science Foundation of China (U21B2032)
- National Key Research and Development Program of China (2023YFB2806802,2021YFB3600104)
This publication has 21 references indexed in Scilit:
- Thermal-Aware Task-Mapping Method for 3D-NoC-Based Neuromorphic Systems2023
- AOME: Autonomous Optimal Mapping Exploration Using Reinforcement Learning for NoC-based Accelerators Running Neural Networks2022
- HotCluster: A Thermal-Aware Defect Recovery Method for Through-Silicon-Vias Toward Reliable 3-D ICs SystemsIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2021
- ConfuciuX: Autonomous Hardware Resource Assignment for DNN Accelerators using Reinforcement Learning2020
- A NoC-based simulator for design and evaluation of deep neural networksMicroprocessors and Microsystems, 2020
- Energy Optimization in 3D Networks-on-Chip through Dynamic Voltage Scaling technique2020
- Tianjic: A Unified and Scalable Chip Bridging Spike-Based and Continuous Neural ComputationIEEE Journal of Solid-State Circuits, 2020
- Eyeriss v2: A Flexible Accelerator for Emerging Deep Neural Networks on Mobile DevicesIEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2019
- Neu-NoC: A high-efficient interconnection network for accelerated neuromorphic systems2018
- Proactive temperature balancing for low cost thermal management in MPSoCs2008