《電子技術(shù)應(yīng)用》
您所在的位置:首頁 > 模擬設(shè)計 > 設(shè)計應(yīng)用 > Innovus機器學(xué)習(xí)在高性能CPU設(shè)計中的應(yīng)用
Innovus機器學(xué)習(xí)在高性能CPU設(shè)計中的應(yīng)用
2020年電子技術(shù)應(yīng)用第8期
邊少鮮1,,Micheal Feng1,David Yue1,,欒曉琨1,,蔡 準2,蔣劍鋒1
1.天津飛騰信息技術(shù)有限公司,,湖南 長沙410000,;2.上海楷登電子科技有限公司,,上海201204
摘要: 高性能芯片設(shè)計在7 nm及更高級的工藝節(jié)點上,,設(shè)計規(guī)模更大、頻率更高,、設(shè)計數(shù)據(jù)和可變性更復(fù)雜,物理設(shè)計難度增大,。機器學(xué)習(xí)在多領(lǐng)域均獲得成功應(yīng)用,,復(fù)雜的芯片設(shè)計是應(yīng)用機器學(xué)習(xí)的一個很好的領(lǐng)域。Cadence將機器學(xué)習(xí)算法內(nèi)置到Innovus工具中,,通過對芯片設(shè)計數(shù)據(jù)進行學(xué)習(xí)建模,,建立機器學(xué)習(xí)模型,從而提升芯片性能表現(xiàn),。建立了一個應(yīng)用機器學(xué)習(xí)優(yōu)化延時的物理流程來提升芯片設(shè)計性能,。詳細討論分析了分別對單元延時、線延時,、單元和線延時進行優(yōu)化對設(shè)計的影響,,進而找到一個較好的延時優(yōu)化方案。最后利用另一款設(shè)計難度更大,,性能要求更高的模塊從時序,、功耗、線長等方面較為全面地分析驗證設(shè)計方案的合理性,。
中圖分類號: TN402
文獻標識碼: A
DOI:10.16157/j.issn.0258-7998.209801
中文引用格式: 邊少鮮,,Micheal Feng,David Yue,,等. Innovus機器學(xué)習(xí)在高性能CPU設(shè)計中的應(yīng)用[J].電子技術(shù)應(yīng)用,,2020,46(8):54-59,63.
英文引用格式: Bian Shaoxian,,Micheal Feng,,David Yue,et al. Innovus machine learning application in performance CPU design[J]. Application of Electronic Technique,,2020,,46(8):54-59,63.
Innovus machine learning application in performance CPU design
Bian Shaoxian1,,Micheal Feng1,,David Yue1,Luan Xiaokun1,,Cai Zhun2,,Jiang Jianfeng1
1.Tianjin Phytium Technology Co.,Ltd.,,Changsha 410000,,China;2.Cadence Design Systems,,Inc.,,Shanghai 202014,China
Abstract: The high-performance chip design has a larger design scale, higher frequency, more complex design data and reliability, and more signoff indicators under 7 nm and higher process nodes. Machine learning has been successfully applied in many fields, and complex chip design is a good field for applying machine learning. Cadence built the algorithm into the Innovus tool, and built the machine learning model by learning and modeling the chip design data to improve chip performance. A physical design process that applies machine learning to optimize latency is established to improve chip design performance. This paper presents a machining-learning-based physical design flow that optimizes delay to improve chip design performance. In orde to choose a better solution,,the effect of optimizing the cell delay,,net delay,cell and net delay separately on the design was discussed and analyisised in detail. Finally,,the solution is applied to another block design with more difficult design and higher performance requirements . To verifies the consistency of the flow,,a more comprehensive analysis is completed from the aspects of timing,power,,wire length,,etc.
Key words : machine learning;Innovus,;chip design,;physical design

0 引言

    摩爾定律揭示了集成電路的集成度和技術(shù)節(jié)點的飛速發(fā)展,這使得芯片設(shè)計的復(fù)雜度和數(shù)據(jù)量快速上升,,尤其是芯片的物理設(shè)計更是涉及海量的數(shù)據(jù)和信息,,且運行時間和設(shè)計周期漫長,迭代一次的時間和資源代價很大,,這對設(shè)計師的經(jīng)驗與能力要求很高,。機器學(xué)習(xí)如今在各個領(lǐng)域都有廣泛的應(yīng)用,其能學(xué)習(xí)數(shù)據(jù)規(guī)律建立模型從而快速推斷結(jié)果[1],。如果能在物理設(shè)計中應(yīng)用機器學(xué)習(xí)挖掘設(shè)計規(guī)律,,且基于推斷的求解來進行物理設(shè)計,,可加速芯片設(shè)計。國內(nèi)外很多學(xué)者在此方面有了成功的研究,,包括PAN D Z等詳細介紹的在物理設(shè)計中應(yīng)用機器學(xué)習(xí)[2],。LI B使用機器學(xué)習(xí)由全局布線線預(yù)測詳細布線結(jié)果[3]。TSMC在物理設(shè)計中應(yīng)用機器學(xué)習(xí)的兩款芯片分別可使頻率提升40 MHz和減少20 000時鐘門控單元等[4],。

    本文基于Cadence Innovus工具建立應(yīng)用機器學(xué)習(xí)進行延時優(yōu)化的物理設(shè)計流程,,研究7 nm工藝下不同層金屬的特性,設(shè)置三個實驗組單元延時優(yōu)化,、線延時優(yōu)化,、單元和線延時同時優(yōu)化與傳統(tǒng)物理設(shè)計流程進行對比分析。同時將應(yīng)用機器學(xué)習(xí)進行延時優(yōu)化的物理設(shè)計流程應(yīng)用到更大規(guī)模,,設(shè)計復(fù)雜度更高的ARM架構(gòu)的一款CPU設(shè)計中,,均得到了很好地性能優(yōu)化。最終確定了兩款模塊芯片均采用Innovus機器學(xué)習(xí)進行延時優(yōu)化的物理設(shè)計流程,。




本文詳細內(nèi)容請下載:http://forexkbc.com/resource/share/2000002945




作者信息:

邊少鮮1,,Micheal Feng1,David Yue1,,欒曉琨1,,蔡  準2,蔣劍鋒1

(1.天津飛騰信息技術(shù)有限公司,,湖南 長沙410000,;2.上海楷登電子科技有限公司,,上海201204)

此內(nèi)容為AET網(wǎng)站原創(chuàng),,未經(jīng)授權(quán)禁止轉(zhuǎn)載,。