隨著半導體製程的進展，電晶體的尺寸也不斷的縮小。可靠度成為一個不可忽視的重要問題。其中，負偏壓溫度不穩定性是一個常見且影響很大的可靠度問題。負偏壓溫度不穩定性會使得電晶體的閾值電壓產生飄移，這可能導致時序違規甚至系統故障。許多先前的研究者提出多種技術來減輕負偏壓溫度不穩定性所帶來的影響。但是，這些技術大多應用於同質多核心系統，無法直接應用在異質多核心系統之上。相較於同質多核心系統皆採用相同架構的核心，異質多核心系統可以在高效能與低功耗的核心之間進行選擇，具有較高的彈性同時有更多的設計空間。
我們透過以上的觀察，在本篇論文中研究了異質多核系統的特性，提出了一種可感知負偏壓溫度不穩定性的框架以提高系統壽命，並且針對不同核心種類採取對應的老化策略。我們針對高效能核心採用不對稱老化方法，此概念可以使一些核心保持健康狀態，以在系統後期處理緊急時限工作。而對於低功耗核心我們採用對稱式老化方法，讓每個核心的老化狀況趨於一致。這樣可以使用更多的低功耗核心共同處理非時限緊急的工作。而工作遷移技術則可以使用不同類型的核心執行單個工作，在功耗與系統壽命之間取得平衡。實驗結果顯示，我們所提出的框架與磨損感知的方法相比，可以實現5.29倍至10.78倍的使用壽命提高，並節省平均功耗11.8％至23.8％。

With the progress of process technology, the size of transistors has been continuously reduced. Reliability has become a non-negligible issue. The Negative Bias Temperature Instability (NBTI) effect is one of the major reliability threats. The NBTI effect will cause the threshold voltage of the transistor to drift, which may lead to timing violations and even system failures. Many previous researchers have proposed multiple techniques to mitigate the impact of the NBTI effect. However, most of these proposed techniques are applied to homogeneous multicore systems. The heterogeneous multicore systems can trade-off between high-performance big core and low-power LITTLE cores, which is more flexible than the homogeneous multicore systems.
Through the above observations, in this thesis, we have studied the characteristics of the heterogeneous multicore systems and proposed an NBTI-Aware Task Scheduling Framework (NETS) to improve the system lifetime and adopted corresponding aging strategies for different core types. We propose to adopt an asymmetric aging approach for high-performance big cores. It can reserve some robust cores to deal with critical tasks in the later life stage of the system. For low-power LITTLE cores, we proposed to adopt a symmetric aging approach, which can balance the aging rate of each core. It can keep more low-power LITTLE cores to deal with non-critical tasks. Finally, we propose a task migration mechanism to execute a task on different types of cores. It further trade-off between power consumption and system lifetime. Experimental results show that the proposed framework can achieve a 5.29 times to 10.78 times longer lifetime and save an average power consumption of 11.8\% to 23.8\% when compared to the wearout-aware method.

[1] Predictive technology model (ptm). In http://www.ptm.asu.edu.
[2] M. A. Alam and S. Mahapatra. A comprehensive model of pmos nbti degradation. In Microelectronics Reliability, vol. 45, issue 1, pages 71–81, 2005.
[3] ARM. big.little processing technologies. In https://www.arm.com/whyarm/technologies/big-little.
[4] Alessandro Baldassari, Cristiana Bolchini, and Antonio Miele. A dynamic reliability management framework for heterogeneous multicore systems. In 2017 IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFT), pages 1–6, 2017.
[5] M. Basoglu, M. Orshansky, and M. Erez. Nbti-aware dvfs: A new approach to saving energy and increasing processor lifetime. In ACM/IEEE international symposium on Low power electronics and design(ISLPED), pages 253–258, Aug 2010.
[6] Hoon Sung Chwa, Jaebaek Seo, Hyuck Yoo, Jinkyu Lee, and Insik Shin. Energy and feasibility optimal global scheduling framework on big. little platforms. In IEEE RTSOPS, pages 1–11, 2014.
[7] K. Van Craeynest, A. Jaleel, L. Eeckhout, P. Narvaez, and J. Emer. Scheduling heterogeneous multi-cores through performance impact estimation (pie). In 2012 39th Annual International Symposium on Computer Architecture (ISCA), pages 213–224, 2012.
[8] R. P. Dick, D. L. Rhodes, and W. Wolf. Tgff: task graphs for free. In International Workshop on Hardware/Software Codesign, pages 97–101, 1998.
[9] M. Ebrahimi, F. Oboril, S. Kiamehr, and M. B. Tahoori. Aging-aware logic synthesis. In 2013 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), pages 61–68, 2013.
[10] D. Gnad, M. Shafique, F. Kriebel, S. Rehman, Duo Sun, and J. Henkel. Hayat: Harnessing dark silicon and variability for aging deceleration and balancing. In 2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC), pages 1–6, 2015.
[11] Sadeque Hamdan and Ali Cheaitou. Supplier selection and order allocation with green criteria: An mcdm and multi-objective optimization approach. In Computers Operations Research, vol. 81, pages 282–304, 2017.
[12] N. K. Jha, P. S. Reddy, D. K. Sharma, and V. R. Rao. Nbti degradation and its impact for analog circuit reliability. In IEEE Transactions on Electron Devices, vol. 52, issue 12, pages 2609–2615, 2005.
[13] P. Mercati, F. Paterna, A. Bartolini, L. Benini, and T. S. Rosing. Dynamic variability management in mobile multicore processors under lifetime constraints. In 2014 IEEE 32nd International Conference on Computer Design (ICCD), page 448–455, Oct 2014.
[14] T. R. Mück, Z. Ghaderi, N. D. Dutt, and E. Bozorgzadeh. Exploiting heterogeneity for aging-aware load balancing in mobile platforms. In IEEE Transactions on Multi-Scale Computing Systems, volume 3, pages 25–35, 2017.
[15] F. Paterna, A. Acquaviva, and L. Benini. Aging-aware energy-efficient workload allocation for mobile multimedia platforms. In IEEE Transactions on Parallel and Distributed Systems, 24(8):1489–1499, Aug 2013.
[16] B. C. Paul, K. Kang, H. Kufluoglu, M. A. Alam, and K. Roy. Negative bias temperature instability: Estimation and design for improved reliability of nanoscale circuits. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 26(4):743–751, 2007.
[17] M. Pricopi, T. S. Muthukaruppan, V. Venkataramani, T. Mitra, and S. Vishin. Powerperformance modeling on asymmetric multi-cores. In 2013 International Conference on Compilers, Architecture and Synthesis for Embedded Systems (CASES), pages 1–10, 2013.
[18] W. J. Song, S. Mukhopadhyay, and S. Yalamanchili. Amdahl’s law for lifetime reliability scaling in heterogeneous multicore processors. In 2016 IEEE International Symposium on High Performance Computer Architecture (HPCA), pages 594–605, 2016.
[19] Jin Sun, Rui Zheng, Jyothi Velamala, Yu Cao, Roman Lysecky, Karthik Shankar, and Janet Roveda. A self-tuning design methodology for power-efficient multi-core systems. In ACM Transactions on Design Automation of Electronic Systems (TODAES), 18(1), 1-24., January 2013.
[20] A. Tiwari and J. Torrellas. Facelift: Hiding and slowing down aging in multicores. In 2008 41st IEEE/ACM International Symposium on Microarchitecture, page 129–140, Nov 2008.
[21] E. Vasilakis, I. Sourdis, V. Papaefstathiou, A. Psathakis, and M. G. H. Katevenis. Modeling energy-performance tradeoffs in arm big.little architectures. In 2017 27th International Symposium on Power and Timing Modeling, Optimization and Simulation (PATMOS), pages 1–8, 2017.
[22] R. Vattikonda, Wenping Wang, and Yu Cao. Modeling and minimization of pmos nbti effect for robust nanometer design. In ACM/IEEE Design Automation Conference (DAC), pages 1047–1052, 2006.
[23] Wenping Wang, Zile Wei, Shengqi Yang, and Yu Cao. An efficient method to identify critical gates under circuit aging. In IEEE/ACM International Conference on Computer- Aided Design (ICCAD), pages 735–740, 2007.