By combining Three Dimensional Integrated Circuits with the Network-on-Chip infrastructure to obtain 3D Networks-on-Chip (3D-NoCs), the new on-chip communication paradigm brings several advantages on lower power, smaller footprint and lower latency. However, thermal dissipation is one of the most critical challenges for 3D-ICs where the heat cannot easily transfer through several layers of silicon. Consequently, the high-temperature area also confronts the reliability threat as the Mean Time to Failure (MTTF) decreases exponentially with the operating temperature. Apparently, 3D-NoCs must tackle this fundamental problem in order to be widely used. Therefore, in this work, we investigate the thermal distribution and reliability prediction of 3D-NoCs. We first present a new method to help simulate the temperature (both steady and transient) using traffics value from realistic and synthetic benchmarks and the power consumption from standard VLSI design flow. Then, based on the proposed method, we further predict the relative reliability between different parts of the network. Experimental results show that the method has an extremely fast execution time in comparison to the acceleration lifetime test. Furthermore, we compare the thermal behavior and reliability between Monolithic design and TSV-based TSV. We also explorer the ability to implement the thermal via a mechanism to help reduce the operating temperature.

中文翻译：

---

3D 片上网络的功率、热量和可靠性预测报告

通过将三维集成电路与片上网络基础设施相结合以获得 3D 片上网络 (3D-NoC)，新的片上通信范式在更低功耗、更小尺寸和更低延迟方面带来了多项优势。然而，散热是 3D-IC 面临的最关键挑战之一，因为热量不容易通过多层硅传递。因此，高温区域也面临可靠性威胁，因为平均故障时间 (MTTF) 随工作温度呈指数下降。显然，3D-NoC 必须解决这个基本问题才能得到广泛应用。因此，在这项工作中，我们研究了 3D-NoC 的热分布和可靠性预测。我们首先提出了一种新方法，使用来自现实和综合基准的流量值以及来自标准 VLSI 设计流程的功耗来帮助模拟温度（稳态和瞬态）。然后，基于所提出的方法，我们进一步预测了网络不同部分之间的相对可靠性。实验结果表明，与加速寿命测试相比，该方法具有极快的执行时间。此外，我们比较了单片设计和基于 TSV 的 TSV 之间的热行为和可靠性。我们还探索了通过一种机制来实现散热的能力，以帮助降低工作温度。我们进一步预测了网络不同部分之间的相对可靠性。实验结果表明，与加速寿命测试相比，该方法具有极快的执行时间。此外，我们比较了单片设计和基于 TSV 的 TSV 之间的热行为和可靠性。我们还探索了通过一种机制来实现散热的能力，以帮助降低工作温度。我们进一步预测了网络不同部分之间的相对可靠性。实验结果表明，与加速寿命测试相比，该方法具有极快的执行时间。此外，我们比较了单片设计和基于 TSV 的 TSV 之间的热行为和可靠性。我们还探索了通过一种机制来实现散热的能力，以帮助降低工作温度。