考虑温度/功耗/热导之间相互作用的单循环迭代热分析算法

doi:10.3969/j.issn.0372-2112.2016.06.006

电子学报 ›› 2016, Vol. 44 ›› Issue (6): 1300-1306.DOI: 10.3969/j.issn.0372-2112.2016.06.006

考虑温度/功耗/热导之间相互作用的单循环迭代热分析算法

潘月斗^1,2, 王嘉琪^1,2, 唐亮³, 骆祖莹³

1. 北京科技大学自动化学院, 北京 100083;
2. 北京科技大学钢铁流程先进控制教育部重点实验室, 北京 100083;
3. 北京师范大学信息科学与技术学院, 北京 100875

收稿日期:2014-07-22 修回日期:2015-06-25 出版日期:2016-06-25
- 作者简介:
- 潘月斗 男,1966年出生,博士,副教授.主要从事交流电动机智能控制理论,研究及高速高精交流电动机驱动系统的计算机数字控制系统设计.E-mail:ydpan@ustb.edu.cn;王嘉琪 男,1989年出生,硕士研究生.主要从事电热综合分析、高性能并行计算方面的研究工作.E-mail:3706157@qq.com
- 基金资助:
- 国家自然科学基金 (No.51331002)

TPG_Sli:Single-Loop Iterative Thermal Analysis Algorithm Considering Interactions Among Temperature,Power and Heat Conductance

PAN Yue-dou^1,2, WANG Jia-qi^1,2, TANG Liang³, LUO Zu-ying³

1. School of Automation, University of Science and Technology Beijing, Beijing 100083, China;
2. Key Laboratory of Advanced Control of Iron and Steel Process (Ministry of Education), Beijing 100083, China;
3. College of Information Science and Technology, Beijing Normal University, Beijing 100875, China

Received:2014-07-22 Revised:2015-06-25 Online:2016-06-25 Published:2016-06-25
- Supported by:
- National Natural Science Foundation of China (No.51331002)

摘要/Abstract

摘要：

随着纳米工艺的不断改进,温度对漏电流功耗和热导的影响日益显著.考虑温度/功耗/热导相互作用的3D芯片热分析需要采用迭代方法对温度进行精确求解,即先用功耗密度向量和热导矩阵来求解温度向量,再用求解出来的温度向量来刷新功耗密度向量和热导矩阵.为了提高3D芯片热分析的效率,本文以一个设定温度值下的均匀热导矩阵作为预条件,先提出了一种双循环、内循环低迭代次数的高效求解算法TPG-FTCG.鉴于TPG-FTCG具有超快的内循环收敛速度,本文省去了TPG-FTCG算法的内循环部分,提出了一种单循环、低迭代次数的TPG求解算法TPG-Sli.基于GPU(Graphics Processing Unit)并行加速技术,本文编写并改进了TPG-Sli的GPU加速算法.实验数据表明:与采用经典高效的ICCG算法进行3D芯片热分析的TPG-ICCG算法相比,在足够小的误差范围内,TPG-Sli的GPU加速算法可以获得120倍的速度提升.

关键词: 算法, 热分析, 快速傅里叶变换, GPU并行

Abstract:

With the improvement of the nanometer technology, the influences among temperature, leakage power and heat conductance become increasingly significant and it should be taken into account in 3D chip comprehensive thermal analysis to solve the accurate temperature based on the iterative solution.The comprehensive thermal analysis method uses the nodal power density vector and the heat conductance matrix to solve the nodal temperature vector, and then, refreshes power density and heat conductance with the obtained nodal temperature.In order to improve the efficiency of 3D chip comprehensive thermal analysis, this work uses the heat conductance matrix as the precondition under a setting temperature.Then it proposes an efficient algorithm TPG-FTCG (CG with the Fast Transform-based Preconditioner) which has double-loop and lower inner-loop iterations.According to TPG-FTCG's fast inner-loop convergence rate, this work removes TPG-FTCG's inner-loop part then proposes a more efficient TPG solving algorithm TPG-Sli (Single-loop iterative), which only has single-loop iterative and fewer iterations.Based on the GPU parallel computing, this work compiles and refines TPG-Sli's GPU-parallel-computing algorithm.Experimental results demonstrate that:On the premise of precision losing, the TPG-Sli's GPU algorithm can achieve about 120X speedup compared with the TPG-ICCG algorithm, which uses the classical and efficient ICCG to deal with the 3D chip comprehensive thermal analysis.

Key words: algorithm, thermal analysis, Fast Fourier transform, GPU parallel computing

中图分类号:

TP393

潘月斗, 王嘉琪, 唐亮, 等. 考虑温度/功耗/热导之间相互作用的单循环迭代热分析算法[J]. 电子学报, 2016, 44(6): 1300-1306.

PAN Yue-dou, WANG Jia-qi, TANG Liang, et al. TPG_Sli:Single-Loop Iterative Thermal Analysis Algorithm Considering Interactions Among Temperature,Power and Heat Conductance[J]. Acta Electronica Sinica, 2016, 44(6): 1300-1306.

参考文献

[1] Liao W P,et al.Temperature and supply voltage aware performance and power modeling at microarchitecture level[J].IEEE Transactions on CAD,2005,24(7):1042-1053.
[2] Liu Z,Tan S X D,et al.Compact nonlinear thermal modeling of packaged integrated systems[A].Proceedings of ASP-DAC[C].Yokohama:IEEE,2013.157-162.
[3] Wu X H,et al.Area minimization of power distribution network using efficient nonlinear programming techniques[J].IEEE Transactions on CAD,2004,23(7):1086-1094.
[4] Daloukas K,et al.A parallel fast transform-based preconditioning approach for electrical-thermal co-simulation of power delivery networks[A].Automation & Test in Europe Conference & Exhibition[C].Grenoble:IEEE,2013.1689-1694.
[5] Zhan Y,Goplen B,et al.Electro-thermal analysis and optimization techniques for nano-scale integrated circuits[A].Proceedings of ASP-DAC[C].Yokohama:IEEE,2006.219-222.
[6] Li P,Pileggi L T,et al.IC thermal simulation and modeling via efficientmultigrid-based approaches[J].IEEE Transactions on CAD,2006,25(9):319-326.
[7] Wang T Y,Lee Y M,et al.3D thermal-ADI-an efficient chip-level transient thermal simulator[A].Proceedings of ISPD[C].Monteray:IEEE,2003.10-17.
[8] Luo Z Y,Tan S X D,et al.Localized statistical 3D thermal analysis considering electro-thermal coupling[A].Proceedings of ISCAS[C].Taibei:IEEE,2009.1289-1292.
[9] Zhan Y,Sapatnekar S S.High efficiency Green function-based thermal simulation algorithms[J].IEEE Transactions on CAD,2007,26(9):1661-1675.
[10] Oh D K,Chen C P,et al.3DFFT:Thermal analysis of non-homogeneous IC using 3D FFT Green function method[A].Proceedings of ISQED[C].San Jose:IEEE,2007.567-573.
[11] Qian H F,Sapatnekar S S,et al.Fast poisson solvers for thermal analysis[A].Computer-Aided Design[C].Monterey:IEEE,2010.698-702.
[12] Lee Y M,Wu T H,et al.A hybrid numerical and analytical thermal simulator for 3-D ICs[A].Proceedings of DATE[C].Grenoble:IEEE,2013.1379-1384.
[13] Yan J Q,Luo Z Y,et al.Accurate architecture-level thermal analysis methods for MPSoC with consideration for leakage power dependence on temperature[A].Proceedings of ISQED[C].Santa Clara:IEEE,2013.178-183.
[14] Feng Z,Li P.Fast thermal analysis on GPU for 3D-ICs with integrated microchannel cooling[A].Proceedings of ICCAD[C].San Jose:IEEE,2010.551-555.
[15] Huang K,Yang X,et al.Efficient electro-thermal co-analysis on CPU+GPU heterogeneous architecture[A].Proceedings of ISQED[C].Santa Clara:IEEE,2012.364-369.
[16] Chen J,Wu X J,Cai R.Parallel processing for accelerated mean shift algorithm with GPU[J].Journal of Computer-Aided Design & Computer Graphics,2010,03:461-466.

考虑温度/功耗/热导之间相互作用的单循环迭代热分析算法

TPG_Sli:Single-Loop Iterative Thermal Analysis Algorithm Considering Interactions Among Temperature,Power and Heat Conductance

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	饶伟, 贾凤勤, 李旦. 有色噪声下不相关和相干混合信号的DOA估计[J]. 电子学报, 2023, 51(3): 622-631.
[2]	李杨帅, 彭斐, 韩倩, 李小帅, 解光军. 一种针对QCA电路自动布局布线的混合策略研究[J]. 电子学报, 2023, 51(3): 666-674.
[3]	刘云涛, 申泽生, 方硕, 王云. 高吞吐率流水线结构的ZUC-256流密码硬件设计[J]. 电子学报, 2023, 51(2): 438-445.
[4]	胡向东, 吕高飞, 白银. 基于优化支持向量回归的工业互联网安全态势预测方法[J]. 电子学报, 2023, 51(2): 446-454.
[5]	熊小峰, 黄淳岚, 乐光学, 戴亚盛, 杨晓慧, 杨忠明. 边缘计算中基于综合信任评价的任务卸载策略[J]. 电子学报, 2022, 50(9): 2134-2145.
[6]	陇盛, 陶蔚, 张泽东, 陶卿. 非光滑强凸情形Adam型算法的最优收敛速率[J]. 电子学报, 2022, 50(9): 2049-2059.
[7]	魏振春, 傅宇, 马仲军, 吕增威, 石雷, 张本宏. 带时间窗的无线可充电传感器网络多目标路径规划算法[J]. 电子学报, 2022, 50(8): 1819-1829.
[8]	宣志玮, 毛剑琳, 张凯翔. CBS框架下面向复杂地图的低拓展度A*算法[J]. 电子学报, 2022, 50(8): 1943-1950.
[9]	钟沛龙, 黎明, 何超, 陈昊. 基于SOM聚类和自适应算子选择的高维多目标进化算法[J]. 电子学报, 2022, 50(8): 1959-1974.
[10]	穆磊, 王鹏. 量子启发式优化算法的尺度动态调速机制[J]. 电子学报, 2022, 50(7): 1664-1673.
[11]	陈嘉兴, 程杰, 董云玲, 刘志华. 基于弯曲声线和测距修正的水下节点定位算法[J]. 电子学报, 2022, 50(7): 1567-1572.
[12]	唐捷凯, 胡蓉, 钱斌, 金怀平, 向凤红. 混合帝国竞争算法求解带多行程批量配送的多工厂集成调度问题[J]. 电子学报, 2022, 50(7): 1621-1630.
[13]	许皓文, 陆浩, 王振占. 高光谱微波辐射计系统中2GHz带宽数字谱仪设计[J]. 电子学报, 2022, 50(6): 1472-1479.
[14]	吴晓晓, 李刚强, 张胜利. 分布式协作频谱感知网络中恶意节点检测和定位方法研究[J]. 电子学报, 2022, 50(6): 1370-1380.
[15]	焦志鹏, 姚富, 陈华, 王舰, 匡晓云, 黄开天. 一种基于门限与感染技术的SM4算法综合防护实现[J]. 电子学报, 2022, 50(5): 1066-1074.