基於圖神經網絡（GNN）的內部節點控制（INC）和輸入向量控制（IVC）協同優化用於老化緩解

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：140

、訪客IP：3.133.135.8

姓名

楊修宜(Hsiu-Yi Yang) 查詢紙本館藏

畢業系所

電機工程學系

論文名稱

基於圖神經網絡（GNN）的內部節點控制（INC）和輸入向量控制（IVC）協同優化用於老化緩解
(Graph Neural Network (GNN) Based Internal Node Control (INC) and Input Vector Control (IVC) Co-Optimization for Aging Mitigation)

相關論文

★ 晶圓圖之網格及稀疏缺陷樣態辨識	★ 晶圓圖提取特徵參數錯誤樣態分析
★ 使用聚類過濾策略和 CNN 計算識別晶圓圖瑕疵樣態	★ 新建晶圓圖相似性門檻以強化相似程度辨別能力
★ 一種可動態重新配置的4:2近似壓縮器用於補償老化	★ 一個可靠的靜態隨機存取記憶體內運算結構: 設計指南與耐老化策略研究
★ 一個高效的老化偵測器部屬策略: 基於生成對抗網路的設計方法	★ 考慮電壓衰退和繞線影響以優化電路時序之電源供應網絡精煉策略
★ 適用於提高自旋轉移力矩式磁阻隨機存取記憶體矩陣可靠度之老化偵測與緩解架構設計	★ 8T 靜態隨機存取記憶體之內積運算引擎的老化威脅緩解策略: 從架構及運算角度來提出解決的方法
★ 用於響應穩定性的老化感知平行掃描鏈PUF設計	★ 8T靜態隨機存取記憶體運算的老化檢測和容忍機制：適用於邏輯和 MAC 運算的應用
★ 使用擺置後的設計特徵及極限梯度提升演算法預測繞線後的繞線需求	★ 基於強化學習的晶片佈局規劃的卷積神經網路與圖神經網路融合架構
★ 用於佈線後階段電壓降優化的強化學習框架	★ 多核心系統的老化與瞬態錯誤感知任務部署策略：壽命延長且節能的框架

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 (2029-7-10以後開放)

摘要(中)

隨著半導體製程的進步，電路老化的問題變得日益突出。其中一種最嚴重的老化效應是負偏溫度不穩定性（NBTI），它會增加PMOS晶體管的閾值電壓和傳播延遲。增加的延遲在長時間下可能導致時序違規，甚至系統故障。為了改善NBTI帶來的問題，提出了一些老化緩解方法，如內部節點控制（INC）和輸入向量控制（IVC）。INC應用特殊設計的邏輯門來緩解老化效應，而IVC則在電路閒置期間使用適當的輸入向量來防止NBTI造成的壓力。然而，INC會導致額外的面積成本和功耗，而電路的結構會限制了IVC的可控性。儘管過去提出了各種基於INC或IVC的老化容忍方法，但只有少數作者有考慮兩種方法之間的協同優化。在本文中，我們介紹了一種基於GNN的INC和IVC協同優化框架，充分利用這兩種方法的優勢，盡可能地降低老化引起的延遲。我們的做法是使用經過良好訓練的GNN來識別電路中受到嚴重老化影響的邏輯閘，然後在給定面積限制下使用INC和IVC來緩解老化效應。實驗結果表示，在電路中採用我們的方法，與先前的研究成果相比，我們可以降低因老化造成額外的傳遞時間延遲大約2.16倍，同時，我們與先前的研究相較之下，我們只需要29.5%額外的面積消耗。此外，我們的方法執行時間也比先前的做法還要少，加快了2.6倍的速度。

摘要(英)

With the advancement of semiconductor processes, the issue of circuit aging is becoming increasingly prominent. One of the most severe aging effects is negative bias temperature instability (NBTI), which increases the threshold voltage and propagation delay of PMOS transistors. Over time, the increased delay may lead to timing violations and even system failures. To address the problems caused by NBTI, various aging mitigation techniques have been proposed, such as Internal Node Control (INC) and Input Vector Control (IVC). INC applies specially designed logic gates to mitigate aging effects, while IVC utilizes appropriate input vectors during circuit idle periods to alleviate the pressure from NBTI. However, INC incurs additional area costs and power consumption, while the structure of the circuit limits the controllability of IVC. Although various aging tolerance methods based on INC or IVC have been proposed in the past, only a few authors have considered the synergistic optimization between the two methods. In this thesis, we introduce a GNN-based INC and IVC co-optimization framework, leveraging the advantages of both methods to minimize aging-induced delays as much as possible. Our approach involves using a well-trained GNN to identify logic gates in the circuit significantly affected by aging, and then utilizing INC and IVC under given area constraints to mitigate aging effects. Experimental results indicate that adopting our method in the circuit can reduce additional propagation delay caused by aging by approximately 2.16 times compared to previous research results, while requiring only 29.5% additional area consumption compared to previous studies. Furthermore, the execution time of our method is also faster, accelerating by 2.6 times compared to previous approaches.

關鍵字(中)

★ 內部節點控制
★ 輸入向量控制
★ 圖神經網路
★ 負偏壓溫度不穩定性
★ 更換閘門
★ 減緩老化

關鍵字(英)

★ Internal Node Control (INC)
★ Input Vector Control (IVC)
★ Graph Neural Network (GNN)
★ Negative Bias Temperature Instability
★ gate replacement
★ aging mitigation

論文目次

摘要 i
Abstract ii
致謝 iii
Table of Contents iv
Table of Figures vi
Table of Tables viii
Chapter 1 Introduction 1
1.1 Aging Issues 1
1.2 Mitigation Methods 2
1.3 Contributions 4
Chapter 2 Preliminaries 6
2.1 Aging Effect and Aging Model 6
2.2 Aging Mitigation Methods 7
2.2.1 Internal Node Control (INC) 7
2.2.2 Input Vector Control (IVC) 8
2.3 Graph Neural Network (GNN) for Electronic Design Automation (EDA) 9
2.3.1 Graph Neural Network (GNN) 10
2.3.2 GNN-Based Approaches for Mitigating Aging Issues 13
Chapter 3 Problem Formulation 15
3.1 Motivation 15
3.2 Problem Formulation 16
Chapter 4 GNN-Based INC and IVC Co-Optimization Framework for Aging Mitigation 17
4.1 Overall Framework 17
4.2 GNN-based Approach for INC Logic Deployment 18
4.3 GNN-based Co-Optimization with INC and IVC 21
4.4 Iterative GNN-based Co-Optimization with INC and IVC 22
Chapter 5 Experimental Results 25
5.1 Experimental Setting 25
5.2 Comparison of Aging-induced Degradation Improvement Rate (Dimp) 27
5.3 Comparison of Gate Replacement Rate 28
5.4 Comparison of Area Overhead (Ainc) 29
5.5 Comparison of Ratio of Delay Improvement over Area Overhead (Dimp/Ainc) 30
5.6 Comparison of Execution Time 31
5.7 Comparison of Leakage Power 33
Chapter 6 Conclusions 34
Reference 35

參考文獻

[1] W. Qingwu, Y. Maoxiang, Y. Shiqi, Z. Yao, D. Li, and L. Huaguo, "Effective critical gate identification for mitigating circuit aging using gate replacement," in IEEE International Conference on Electronic Measurement & Instruments (ICEMI), Yangzhou, China, pp. 204-208, 2017.
[2] A. Calimera, E. Macii, and M. Poncino, "Design Techniques for NBTI-Tolerant Power-Gating Architectures," in IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 59, no. 4, pp. 249-253, April 2012.
[3] S. M. Ebrahimipour, B. Ghavami, and M. Raji, "A Statistical Gate Sizing Method for Timing Yield and Lifetime Reliability Optimization of Integrated Circuits," in IEEE Transactions on Emerging Topics in Computing, vol. 9, no. 2, pp. 759-773, 1 April-June 2021.
[4] F. Firouzi, F. Ye, K. Chakrabarty, and M. B. Tahoori, "Aging-and variation-aware delay monitoring using representative critical path selection," in ACM Transactions on Design Automation of Electronic Systems (TODAES), vol. 20, no. 3, pp. 1-23, Jun. 2015.
[5] S. Arasu, M. Nourani, J. M. Carulli, and V. K. Reddy, "Controlling Aging in Timing-Critical Paths," in IEEE Design & Test, vol. 33, no. 4, pp. 82-91, Aug. 2016.
[6] S. Gupta and S. S. Sapatnekar, "GNOMO: Greater-than-NOMinal Vdd operation for BTI mitigation," in Asia and South Pacific Design Automation Conference (ASPDAC), Sydney, NSW, Australia, 2012, pp. 271-276.
[7] E. Mintarno, et al., "Self-Tuning for Maximized Lifetime Energy-Efficiency in the Presence of Circuit Aging," in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 30, no. 5, pp. 760-773, May 2011.
[8] Y. Wang, X. Chen, W. Wang, Y. Cao, Y. Xie, and H. Yang, "Leakage Power and Circuit Aging Cooptimization by Gate Replacement Techniques," in IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 19, no. 4, pp. 615-628, April 2011.
[9] Y. Wang, Xiaoming Chen, W. Wang, Y. Cao, Y. Xie, and H. Yang, "Gate replacement techniques for simultaneous leakage and aging optimization," Design, Automation & Test in Europe Conference & Exhibition, Nice, France, pp. 328-333, 2009.
[10] David R. Bild, Robert P. Dick, and Gregory E. Bok, "Static NBTI Reduction Using Internal Node Control," in ACM Transactions on Design Automation of Electronic Systems (TODAES), vol. 17, no. 4, pp. 1-30, October 2012.
[11] D. R. Bild, G. E. Bok, and R. P. Dick, "Minimization of NBTI performance degradation using internal node control," Design, Automation & Test in Europe Conference & Exhibition, Nice, France, pp. 148-153, 2009.
[12] G. Zhang, M. Yi, Y. Miao, D. Xu, and H. Liang, "NBTI-induced circuit aging optimization by protectability-aware gate replacement technique," Latin-American Test Symposium (LATS), Puerto Vallarta, Mexico, 2015, pp. 1-4.
[13] Y. Wang, H. Luo, K. He, R. Luo, H. Yang, and Y. Xie, "Temperature-Aware NBTI Modeling and the Impact of Standby Leakage Reduction Techniques on Circuit Performance Degradation," in IEEE Transactions on Dependable and Secure Computing, vol. 8, no. 5, pp. 756-769, Sept.-Oct. 2011.
[14] S. Pendyala, S. A. Islam, and S. Katkoori, "Gate Level NBTI and Leakage Co-Optimization in Combinational Circuits with Input Vector Cycling," in IEEE Transactions on Emerging Topics in Computing, vol. 8, no. 3, pp. 738-749, 1 July-Sept. 2020.
[15] Sun, Peng, Zhiming Yang, Yang Yu, Junbao Li, and Xiyuan Peng, "NBTI and Power Reduction Using an Input Vector Control and Supply Voltage Assignment Method," Algorithms 10, no. 3: 94, 2017.
[16] Wang, Yu, et al., "On the efficacy of input vector control to mitigate NBTI effects and leakage power," in International Symposium on Quality Electronic Design (ISQED), San Jose, CA, USA, pp. 19-26, 2009.
[17] F. Firouzi, S. Kiamehr and M. B. Tahoori, "Power-Aware Minimum NBTI Vector Selection Using a Linear Programming Approach," in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 32, no. 1, pp. 100-110, Jan. 2013.
[18] W. Wang, S. Yang, S. Bhardwaj, S. Vrudhula, F. Liu and Y. Cao, "The Impact of NBTI Effect on Combinational Circuit: Modeling, Simulation, and Analysis," in IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 18, no. 2, pp. 173-183, Feb. 2010.
[19] X. Yang, Q. Sang, C. Wang, M. Yu and Y. Zhao, "Development and Challenges of Reliability Modeling From Transistors to Circuits," in IEEE Journal of the Electron Devices Society, vol. 11, pp. 179-189, 2023.
[20] K. Singh and S. Kalra, "A Comprehensive Assessment of Current Trends in Negative Bias Temperature Instability (NBTI) Deterioration," in International Conference on Signal Processing and Communication (ICSC), pp. 271-276, 2021.
[21] Zhou, Jie, et al., "Graph neural networks: A review of methods and applications," in AI open 1: 57-81, 2020.
[22] Y. Zhang, H. Ren and B. Khailany, "GRANNITE: Graph Neural Network Inference for Transferable Power Estimation," in ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, USA, pp. 1-6, 2020.
[23] Bu, Aiguo, and Jie Li. "A learning-based framework for circuit path level NBTI degradation prediction." in Electronics, 9.11: 1976, 2020.
[24] B. Tudor, et al., "MOSRA: An efficient and versatile MOS aging modeling and reliability analysis solution for 45nm and below," in IEEE International Conference on Solid-State and Integrated Circuit Technology, Shanghai, China, pp. 1645-1647, 2010.
[25] A. Paszke, et al., "Pytorch: An imperative style, high-performance deep learning library." Advances in neural information processing systems 32, 2019, [online] Available at: http://papers.neurips.cc/paper/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf
[26] Sánchez, Daniela, et al. "A comprehensive survey on electronic design automation and graph neural networks: Theory and applications." in ACM Transactions on Design Automation of Electronic Systems (TODAES), vol. 28, no. 2, pp. 1-27, 2023.
[27] Ren, Haoxing, et al. "Why are graph neural networks effective for eda problems?." Proceedings of IEEE/ACM International Conference on Computer-Aided Design (ICCAD), 2022.
[28] L. Alrahis, J. Knechtel and O. Sinanoglu, "Graph neural networks: A powerful and versatile tool for advancing design, reliability, and security of ICs." Proceedings of Asia and South Pacific Design Automation Conference. 2023.
[29] L. Alrahis, J. Knechtel, F. Klemme, H. Amrouch and O. Sinanoglu, "GNN4REL: Graph Neural Networks for Predicting Circuit Reliability Degradation," in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 41, no. 11, pp. 3826-3837, Nov. 2022.
[30] Guo, Zizheng, et al. "A timing engine inspired graph neural network model for pre-routing slack prediction." Proceedings of ACM/IEEE Design Automation Conference. 2022.
[31] D. S. Lopera and W. Ecker, "Applying GNNs to Timing Estimation at RTL." Proceedings of IEEE/ACM International Conference on Computer-Aided Design (ICCAD), 2022.
[32] E. Ustun, C. Deng, D. Pal, Z. Li and Z. Zhang, "Accurate Operation Delay Prediction for FPGA HLS Using Graph Neural Networks," Proceedings of IEEE/ACM International Conference on Computer-Aided Design (ICCAD), San Diego, CA, USA, pp. 1-9, 2020.
[33] Y. -C. Lu, S. Nath, V. Khandelwal and S. K. Lim, "RL-Sizer: VLSI Gate Sizing for Timing Optimization using Deep Reinforcement Learning," ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, USA, pp. 733-738, 2021.
[34] Zhou, Xinyi, et al. "Heterogeneous graph neural network-based imitation learning for gate sizing acceleration." Proceedings of IEEE/ACM International Conference on Computer-Aided Design (ICCAD), 2022.

指導教授

陳聿廣(Yu-Guang Chen)

審核日期

2024-7-11

推文