Fast Analysis of Static IR Drop Effect Based on Machine Learning Methods
https://doi.org/10.15514/ISPRAS-2022-35(5)-9
Abstract
As part of the ICCAD Contest 2023 (Problem C) competition, the paper describes a methodology for applying ML models to perform static IR drop analysis. Methods for obtaining a database for training a neural network to solve this problem are given. We consider a technique for training an ML model to analyze the static IR-drop effect. The generation of input data for training a neural network from SPICE netlists is also discussed in this paper. This solution is ranked in the TOP 3 at the ICCAD Contest 2023 competition.
About the Authors
Roman Aleksandrovich SOLOVYEVRussian Federation
Dr. Sci. (Tech.), member-corr. RAS, chief researcher at the Institute of Design Problems in Microelectronics RAS. Area of scientific interests: automation of digital VLSI design, neural networks and machine learning, residue number system, physical synthesis of integrated circuits.
Dmitry Vladimirovich TELPUKHOV
Russian Federation
Dr. Sci. (Tech.), Deputy Director for Scientific Work of the Institute of Design Problems in Microelectronics RAS. Research interests: automation of digital VLSI design, logic synthesis, radiation-resistant design, machine learning and neural networks, residual class system.
Evgeny Denisovich DEMIDOV
Russian Federation
2nd year postgraduate student of National Research University "MIET", development engineer computer-aided design tools of VLSI development department at Alphachip Company. Theme of scientific research work: «Development of tools for analyzing IR drop effect using machine learning methods». Research interests: automation of digital VLSI design, special types of VLSI analysis, machine learning and neural networks.
Ilya Ilyich SHAFEEV
Russian Federation
2nd year master’s student of National Research University "MIET", research-engineer of the Institute of Design Problems in Microelectronics RAS. Research interests: automation of digital VLSI design, special types of VLSI analysis, machine learning and neural networks.
References
1. Интернет-ресурс https://siliconvlsi.com/what-is-ir-drop/ (дата обращения – 27.11.2023).
2. Интернет-ресурс https://vlsi-backend-adventure.com/ir_analysis.html (дата обращения – 27.11.2023).
3. Интернет-ресурс https://teamvlsi.com/2020/07/ir-analysis-in-asic-design-effects-and.html (дата обращения – 27.11.2023).
4. Chhabria V. A., Ahuja V., Prabhu A., Patil N., Jain P., and Sapatnekar S. S. Thermal and IR Drop Analysis Using Convolutional Encoder-Decoder Networks. Proceedings of Asia and South Pacific Design Automation Conference (ASP-DAC), 2021.
5. Chia-Tung Ho and Andrew B Kahng IncPIRD: Fast Learning Based Prediction of Incremental IR Drop. IEEE/ACM International Conference on Computer-Aided Design (ICCAD), 2019.
6. Zhiyao Xie, Haoxing Ren, Brucek Khailany, Ye Sheng, Santosh Santosh, Jiang Hu, and Yiran Chen PowerNet: Transferable Dynamic IR Drop Estimation via Maximum Convolutional Neural Network. Asia and South Pacific Design Automation Conference (ASP-DAC), 2020.
7. Chi-Hsien Pao, An-Yu Su, and Yu-Min Lee XGBIR: An xgboost-based IR drop predictor for power delivery network. Design, Automation & Test in Europe Conference & Exhibition (DATE), 2020
8. Chhabria V. A., Zhang Y., Ren H., Keller B., Khailany B., and Sapatnekar S. S. MAVIREC: ML-Aided Vectored IR-Drop Estimation and Classification. Proceedings of Design, Automation, and Test in Europe (DATE), 2021.
9. Интернет-ресурс https://2023.iccad.com/ (дата обращения – 27.11.2023).
10. Интернет-ресурс http://iccad-contest.org/ (дата обращения – 27.11.2023).
11. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18 (pp. 234-241). Springer International Publishing.
12. Интернет-ресурс https://skine.ru/articles/533011/ (дата обращения – 27.11.2023).
13. Интернет-ресурс https://academy.yandex.ru/handbook/ml/article/metriki-klassifikacii-i-regressii (дата обращения – 27.11.2023).
14. Интернет-ресурс https://eda.ncsu.edu/freepdk/freepdk45/ (дата обращения – 27.11.2023).
15. Интернет-ресурс https://github.com/UMN-EDA/BeGAN-benchmarks (дата обращения – 27.11.2023).
16. Chhabria V. A. et al. BeGAN: Power grid benchmark generation using a process-portable GAN-based methodology. 2021 IEEE/ACM International Conference on Computer Aided Design (ICCAD). – IEEE, 2021. – С. 1-8.
17. Интернет-ресурс https://github.com/The-OpenROAD-Project (дата обращения – 27.11.2023).
18. Ajayi T., Blaauw D. Openroad: Toward a self-driving, open-source digital layout implementation tool chain. Proceedings of Government Microcircuit Applications and Critical Technology Conference. – 2019.
19. Интернет-ресурс https://electroandi.ru/toe/metod/metod-uzlovykh-potentsialov.html (дата обращения 27.11.2023).
20. Интернет-ресурс http://vscripts.ru/w/Схемы_ISCAS85 (дата обращения 27.11.2023).
21. Kirillov, A., He, K., Girshick, R., & Dollár, P. (2017). A unified architecture for instance and semantic segmentation. In CVPR.
22. Tan M., Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks. International conference on machine learning. – PMLR, 2019. – С. 6105-6114.
23. Tu, Z., Talebi, H., Zhang, H., Yang, F., Milanfar, P., Bovik, A., & Li, Y. (2022, October). Maxvit: Multi-axis vision transformer. European conference on computer vision (pp. 459-479). Cham: Springer Nature Switzerland.
24. Shanmugam D. et al. Better aggregation in test-time augmentation //Proceedings of the IEEE/CVF international conference on computer vision. – 2021. – С. 1214-1223.
25. Интернет-ресурс https://ngspice.sourceforge.io/ (дата обращения 27.11.2023).
Review
For citations:
SOLOVYEV R.A., TELPUKHOV D.V., DEMIDOV E.D., SHAFEEV I.I. Fast Analysis of Static IR Drop Effect Based on Machine Learning Methods. Proceedings of the Institute for System Programming of the RAS (Proceedings of ISP RAS). 2023;35(5):127-144. (In Russ.) https://doi.org/10.15514/ISPRAS-2022-35(5)-9