高性能众核处理器芯片时钟网络设计

doi:10.19678/j.issn.1000-3428.0062378

摘要/Abstract

摘要： 随着芯片工艺演进与设计规模增加，高性能众核处理器芯片时钟网络设计面临时序和功耗的全方位挑战。为降低芯片时钟网络功耗并缓解时钟网络分布受片上偏差影响导致的时钟偏斜，在H-Tree+MESH混合时钟网络结构的基础上，结合新一代众核处理器芯片面积大及核心时钟网络分布广的特点，基于标准多源时钟树设计策略构建多源时钟树综合（MRCTS）结构，通过全局H-Tree时钟树保证芯片不同区域间时钟偏斜的稳定可控，利用局部时钟树综合进行关键路径的时序优化以实现时序收敛。实验结果表明，MRCTS能在保证时钟延时、时钟偏斜等性能参数可控的基础上，有效降低时钟网络的负载和功耗，大幅压缩综合子模块的布线资源，加速关键路径的时序收敛，并且在相同电源电压和时钟频率的实测条件下，可获得约22.15%的时钟网络功耗优化。

关键词: 高性能众核处理器芯片, 时钟网络, 时钟功耗, 时钟偏斜, 多源时钟树综合

Abstract: With the evolution of chip technology and an increase in the design scale, the chip clock network design of high-performance many-core processor is facing many challenges, including the power consumption and timing.In order to reduce the power consumption of the chip clock network and the clock skew caused by the On-Chip Variation(OCV) of clock network distribution, this study builds a Multi-Root Clock Tree Synthesis(MRCTS) structure by using the standard multi-source clock tree design strategy.The construction of MRCTS is based on the H-Tree+MESH hybrid clock network structure, and takes into account the large chip area of the new generation many-core processors and the wide distribution of core clock networks.MRCTS ensures stable and controllable clock skew between different regions of the chip through the global H-Tree clock tree and uses local Clock Tree Synthesis(CTS) to optimize the timing of the critical paths and achieve a timing convergence.The experimental results show that MRCTS can effectively reduce the load and power consumption of the clock network by ensuring the controllability of the clock delay, clock skew, and other performance parameters.In addition, it greatly compress the routing resources of each integrated sub-module and accelerates the timing closure of critical paths.Particularly under the measured conditions of the same supply voltage and clock frequency, MRCTS can optimize the clock network power consumption by approximately 22.15%.

Key words: high-performance many-core processor chip, clock network, clock power consumption, clock skew, Multi-Root Clock Tree Synthesis(MRCTS)

中图分类号:

TP393

马永飞, 高成振, 黄金明, 李研. 高性能众核处理器芯片时钟网络设计[J]. 计算机工程, 2022, 48(8): 25-29,36.

MA Yongfei, GAO Chengzhen, HUANG Jinming, LI Yan. Chip Clock Network Design for High-Performance Many-Core Processor[J]. Computer Engineering, 2022, 48(8): 25-29,36.

https://www.ecice06.com/CN/Y2022/V48/I8/25

图/表 11

20220825090705

20220825090709

20220825090716

20220825090720

20220825090724

20220825090728

20220825090732

20220825090736

20220825090740

20220825090747

20220825090752

参考文献

[1] 殷瑞祥, 郭镕, 陈敏.同步数字集成电路设计中的时钟树分析[J].华南理工大学学报(自然科学版), 2005, 33(6):5-8. YIN R X, GUO R, CHEN M.Clock tree analysis in design of synchronous digital integrated circuit[J].Journal of South China University of Technology(Natural Science), 2005, 33(6):5-8.(in Chinese)
[2] WESTE N H, HARRIS D.CMOS超大规模集成电路设计[M].汪东, 译.北京:中国电力出版社, 2006. WESTE N H, HARRIS D.CMOS VLSI design[M].Translated by WANG D.Beijing:China Electric Power Press, 2006.(in Chinese)
[3] 王兵, 彭瑞华, 傅育熙.前后端协同的时钟树设计方法[J].计算机工程, 2008, 34(12):227-229, 232. WANG B, PENG R H, FU Y X.Clock tree design method with front-end and back-end combined[J].Computer Engineering, 2008, 34(12):227-229, 232.(in Chinese)
[4] WANG T J, HUANG S H, CHENG W K, et al.Top-level activity-driven clock tree synthesis with clock skew variation considered[C]//Proceedings of 2016 IEEE International Symposium on Circuits and Systems.Washington D.C., USA:IEEE Press, 2016:2591-2594.
[5] CHAN T B, HAN K, KAHNG A B, et al.OCV-aware top-level clock tree optimization[C]//Proceedings of the 24th Edition of the Great Lakes Symposium on VLSI.New York, USA:ACM Press, 2014:33-38.
[6] TEH E K, ZAWAWI M A M, MOHAMED M F P, et al.Practical full chip clock distribution design with a flexible topology and hybrid metaheuristic technique[J].IEEE Access, 2021, 9:14816-14835.
[7] KAO H Y, LEE Y, HUANG S H, et al.An industrial design methodology for the synthesis of OCV-aware top-level clock tree[C]//Proceedings of the 6th International Symposium on Next Generation Electronics.Washington D.C., USA:IEEE Press, 2017:1-3.
[8] HAN K, KAHNG A B, LI J J.Optimal generalized H-Tree topology and buffering for high-performance and low-power clock distribution[J].IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2020, 39(2):478-491.
[9] VISHNU P V, PRIYARENJINI A R, KOTHA N.Clock tree synthesis techniques for optimal power and timing convergence in SoC partitions[C]//Proceedings of the 4th International Conference on Recent Trends on Electronics, Information, Communication &Technology.Washington D.C., USA:IEEE Press, 2019:276-280.
[10] FRIEDMAN E G.Clock distribution networks in synchronous digital integrated circuits[J].Proceedings of the IEEE, 2001, 89(5):665-692.
[11] LU J W, CHOW W K, SHAM C W.Fast power- and slew-aware gated clock tree synthesis[J].IEEE Transactions on Very Large Scale Integration Systems, 2012, 20(11):2094-2103.
[12] ROY S, MATTHEAKIS P M, MASSE-NAVETTE L, et al.Evolving challenges and techniques for nanometer SoC clock network synthesis[C]//Proceedings of the 12th IEEE International Conference on Solid-State and Integrated Circuit Technology.Washington D.C., USA:IEEE Press, 2014:1-10.
[13] 俞剑明.容偏差百核片上网络设计与多核粒度建模[D].上海:复旦大学, 2014. YU J M.100 core process variation tolerant network on chip design and many-core granularity modeling[D].Shanghai:Fudan University, 2014.(in Chinese)
[14] YEH C, WILKE G, CHEN H, et al.Clock distribution architectures:a comparative study[C]//Proceedings of the 7th International Symposium on Quality Electronic Design.Washington D.C., USA:IEEE Press, 2006:85-91.
[15] SHIH X W, LEE H C, HO K H, et al.High variation-tolerant obstacle-avoiding clock mesh synthesis with symmetrical driving trees[C]//Proceedings of 2010 IEEE/ACM International Conference on Computer-Aided Design.Washington D.C., USA:IEEE Press, 2010:452-457.
[16] RAJARAM A, PAN D Z.Robust chip-level clock tree synthesis[J].IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2011, 30(6):877-890.
[17] SRIVATSA V G, CHAVAN A P, MOURYA D.Design of low power &high performance multi source H-tree clock distribution network[C]//Proceedings of 2020 IEEE VLSI Device Circuit and System.Washington D.C., USA:IEEE Press, 2020:468-473.
[18] XIAO L F, XIAO Z G, QIAN Z C, et al.Local clock skew minimization using blockage-aware mixed tree-mesh clock network[C]//Proceedings of 2010 IEEE/ACM International Conference on Computer-Aided Design.Washington D.C., USA:IEEE Press, 2010:458-462.
[19] 胡向东, 柯希明, 尹飞, 等.高性能众核处理器申威26010[J].计算机研究与发展, 2021, 58(6):1155-1165. HU X D, KE X M, YIN F, et al.Shenwei-26010:a high-performance many-core processor[J].Journal of Computer Research and Development, 2021, 58(6):1155-1165.(in Chinese)
[20] NIKHIL P.A novel clock distribution technology Multisource Clock Tree System(MCTS)[J].International Journal of Advanced Research in Electrical, Electronics and Instrumentation Energy, 2013, 27(6):2234-2239.
[21] BHAT V K, SURENDRA H H, ARCHANA H R.Skew analysis on multisource clock tree synthesis using H-tree structure[M].Berlin, Germany:Springer, 2020.
[22] 高旭.数字后端低功耗设计策略探讨——基于Synopsys EDA工具对时钟树功耗进行分析及优化[J].中国集成电路, 2016, 25(8):31-36, 42. GAO X.Explore the strategy of low power clock tree in physical design-analyze and optimize of clock tree power consumption based on Synopsys EDA tool[J].China Integrated Circuit, 2016, 25(8):31-36, 42.(in Chinese)

选择文件类型/文献管理软件名称

选择包含的内容