New Process to Simultaneously Measure, Quantify, and Model Energy Efficient Performance

M. Mattwandel
{"title":"New Process to Simultaneously Measure, Quantify, and Model Energy Efficient Performance","authors":"M. Mattwandel","doi":"10.1109/MTV.2012.13","DOIUrl":null,"url":null,"abstract":"The term Energy Efficient Performance has been used for over a decade to describe high performance at optimized power levels. While power characterization methods and formal performance assessment techniques are well established and utilized, no standard exists to quantify both metrics simultaneously as a means to establish energy efficient performance. As energy budgets continue to shrink and the expectation for increased performance remains, the need for a process that simultaneously models power versus performance tradeoffs becomes ever more critical. This paper describes a new process developed, prototyped, and implemented in Intel's client post-Si validation labs that is actively being used to measure, process, and model energy efficient performance. The paper will describe methods that simultaneously acquire power and performance measurements, post-process that power data into energy measurements, and model multi-dimensional energy vs. performance tradeoffs. It will also demonstrate how the process is actively being applied to isolate negative energy efficient performance outliers for debug and to tune for best energy efficient performance across multiple product segments.","PeriodicalId":201401,"journal":{"name":"2012 13th International Workshop on Microprocessor Test and Verification (MTV)","volume":"245 1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 13th International Workshop on Microprocessor Test and Verification (MTV)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MTV.2012.13","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

The term Energy Efficient Performance has been used for over a decade to describe high performance at optimized power levels. While power characterization methods and formal performance assessment techniques are well established and utilized, no standard exists to quantify both metrics simultaneously as a means to establish energy efficient performance. As energy budgets continue to shrink and the expectation for increased performance remains, the need for a process that simultaneously models power versus performance tradeoffs becomes ever more critical. This paper describes a new process developed, prototyped, and implemented in Intel's client post-Si validation labs that is actively being used to measure, process, and model energy efficient performance. The paper will describe methods that simultaneously acquire power and performance measurements, post-process that power data into energy measurements, and model multi-dimensional energy vs. performance tradeoffs. It will also demonstrate how the process is actively being applied to isolate negative energy efficient performance outliers for debug and to tune for best energy efficient performance across multiple product segments.
同时测量、量化和模拟节能性能的新过程
术语节能性能已经使用了十多年来描述在优化功率水平下的高性能。虽然功率表征方法和正式的性能评估技术已经很好地建立和使用,但没有标准可以同时量化这两种指标,作为建立节能性能的手段。随着能源预算的不断缩减和对提高性能的期望的持续存在,对同时模拟功率与性能权衡的过程的需求变得越来越重要。本文描述了在英特尔的客户端后si验证实验室中开发、原型化和实现的一个新过程,该过程被积极地用于测量、处理和建模能效性能。本文将描述同时获取功率和性能测量的方法,将功率数据后处理为能量测量,并建立多维能量与性能权衡模型。它还将演示如何积极应用该过程来隔离负能效性能异常值以进行调试,并在多个产品细分市场中调整最佳能效性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信