This repository contains the public materials for the ICCAD 2024 paper ATPlace2.5D. Due to intellectual property restrictions, we are unable to directly provide the core code of ATPlace 🔒. However, this repository includes the following resources:
- Test Cases: A set of test cases used in our experiments.
- Thermal Configuration: The thermal setup used in our work, which is processed using HotSpot. You can compile HotSpot-7.0 on your own 💻.
- Partial Code: A subset of the TAP2.5D codebase 🧩.
In the future, we plan to release a compiled version of the ATPlace code to further support research in this area 🚀.
The test cases are organized into 10 folders (Case1 to Case10), each containing the following files:
- Chiplet Geometric Sizes:
*.blocksfiles that define the geometric sizes of chiplets. - Nets:
*.netsfiles that describe the netlist connectivity. - Pin Locations:
*.plfiles that specify the pin locations. - Chiplet Power:
*.powerfiles that define the power consumption of chiplets.
These files are formatted based on the Bookshelf format, a widely-used standard in placement research. The parsers can be fount in `utils'.
The thermal modeling is performed using HotSpot-7.0, an open-source thermal modeling tool. To use the thermal configuration provided here, follow these steps:
- Clone or download HotSpot-7.0 from its official repository at https://github.com/uvahotspot/HotSpot.git.
- Compile HotSpot-7.0 on your system.
- Use the thermal configuration files provided in this repository as input to HotSpot.
We are committed to open-sourcing more components of ATPlace in the future. Specifically, we plan to release a compiled version of the ATPlace code to enable researchers to reproduce and build upon our work.
- Clone the repository:
git clone https://github.com/Brilight/ATPlace_pub.git
- Explore the test cases in the cases directory.
- Set up HotSpot-7.0 and use the provided thermal configuration files for thermal modeling.
- Results can be compared with the published data in our paper.
If you find this repository useful for your research, please consider citing our ICCAD 2024 paper:
@inproceedings{wang2024atplace2,
title={ATPlace2. 5D: Analytical thermal-aware chiplet placement framework for large-scale 2.5 D-IC},
author={Wang, Qipan and Li, Xueqing and Jia, Tianyu and Lin, Yibo and Wang, Runsheng and Huang, Ru},
booktitle={Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design},
pages={1--9},
year={2024}
}
For questions or feedback, feel free to reach out:
Email: qpwang@pku.edu.cn GitHub Issues: Open an issue in this repository.