基于动态电压和频率缩放的嵌入式gpu运行时功率建模

Q4 Social Sciences Meta: Avaliacao Pub Date : 2020-01-21 DOI:10.1145/3381427.3381429
J. Núñez-Yáñez, Kris Nikov, K. Eder, Mohammad Hosseinabady
{"title":"基于动态电压和频率缩放的嵌入式gpu运行时功率建模","authors":"J. Núñez-Yáñez, Kris Nikov, K. Eder, Mohammad Hosseinabady","doi":"10.1145/3381427.3381429","DOIUrl":null,"url":null,"abstract":"This paper investigates the application of a robust CPU-based power modelling methodology that performs an automatic search of explanatory events derived from performance counters to embedded GPUs. A 64-bit Tegra TX1 SoC is configured with DVFS enabled and multiple CUDA benchmarks are used to train and test models optimized for each frequency and voltage point. These optimized models are then compared with a simpler unified model that uses a single set of model coefficients for all frequency and voltage points of interest. To obtain this unified model, a number of experiments are conducted to extract information on idle, clock and static power to derive power usage from a single reference equation. The results show that the unified model offers competitive accuracy with an average 5% error with four explanatory variables on the test data set and it is capable to correctly predict the impact of voltage, frequency and temperature on power consumption. This model could be used to replace direct power measurements when these are not available due to hardware limitations or worst-case analysis in emulation platforms.","PeriodicalId":38836,"journal":{"name":"Meta: Avaliacao","volume":"17 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2020-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Run-Time Power Modelling in Embedded GPUs with Dynamic Voltage and Frequency Scaling\",\"authors\":\"J. Núñez-Yáñez, Kris Nikov, K. Eder, Mohammad Hosseinabady\",\"doi\":\"10.1145/3381427.3381429\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper investigates the application of a robust CPU-based power modelling methodology that performs an automatic search of explanatory events derived from performance counters to embedded GPUs. A 64-bit Tegra TX1 SoC is configured with DVFS enabled and multiple CUDA benchmarks are used to train and test models optimized for each frequency and voltage point. These optimized models are then compared with a simpler unified model that uses a single set of model coefficients for all frequency and voltage points of interest. To obtain this unified model, a number of experiments are conducted to extract information on idle, clock and static power to derive power usage from a single reference equation. The results show that the unified model offers competitive accuracy with an average 5% error with four explanatory variables on the test data set and it is capable to correctly predict the impact of voltage, frequency and temperature on power consumption. This model could be used to replace direct power measurements when these are not available due to hardware limitations or worst-case analysis in emulation platforms.\",\"PeriodicalId\":38836,\"journal\":{\"name\":\"Meta: Avaliacao\",\"volume\":\"17 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-01-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Meta: Avaliacao\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3381427.3381429\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"Social Sciences\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Meta: Avaliacao","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3381427.3381429","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Social Sciences","Score":null,"Total":0}
引用次数: 2

摘要

本文研究了一种基于cpu的强大功率建模方法的应用,该方法可以自动搜索来自嵌入式gpu的性能计数器的解释性事件。64位Tegra TX1 SoC配置了DVFS功能,并使用多个CUDA基准测试来训练和测试针对每个频率和电压点优化的模型。然后将这些优化模型与一个更简单的统一模型进行比较,该模型对所有感兴趣的频率和电压点使用一组模型系数。为了得到这个统一的模型,我们进行了大量的实验来提取空闲、时钟和静态功耗的信息,从一个参考方程中得出功耗。结果表明,该统一模型对测试数据集具有4个解释变量的平均误差为5%的竞争精度,能够正确预测电压、频率和温度对功耗的影响。当由于硬件限制或仿真平台中的最坏情况分析而无法使用直接功率测量时,该模型可用于替代直接功率测量。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Run-Time Power Modelling in Embedded GPUs with Dynamic Voltage and Frequency Scaling
This paper investigates the application of a robust CPU-based power modelling methodology that performs an automatic search of explanatory events derived from performance counters to embedded GPUs. A 64-bit Tegra TX1 SoC is configured with DVFS enabled and multiple CUDA benchmarks are used to train and test models optimized for each frequency and voltage point. These optimized models are then compared with a simpler unified model that uses a single set of model coefficients for all frequency and voltage points of interest. To obtain this unified model, a number of experiments are conducted to extract information on idle, clock and static power to derive power usage from a single reference equation. The results show that the unified model offers competitive accuracy with an average 5% error with four explanatory variables on the test data set and it is capable to correctly predict the impact of voltage, frequency and temperature on power consumption. This model could be used to replace direct power measurements when these are not available due to hardware limitations or worst-case analysis in emulation platforms.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Meta: Avaliacao
Meta: Avaliacao Social Sciences-Education
CiteScore
0.40
自引率
0.00%
发文量
13
审稿时长
10 weeks
期刊最新文献
Camera spectral sensitivity estimation based on spectrally tunable LED illumination Metamer mismatch volume calculation method based on high-dimensional spherical sampling Machine vision-based portable track inspection system Optimization of RGB image spectral reconstruction based on radial basis function networks Study on spectral adaptive transformation based on chromatic aberration
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1