在mpsoc上使用强制FSMs实现非功能属性的多需求强制-一个案例研究

Khalil Esper, S. Wildermann, J. Teich
{"title":"在mpsoc上使用强制FSMs实现非功能属性的多需求强制-一个案例研究","authors":"Khalil Esper, S. Wildermann, J. Teich","doi":"10.4230/OASIcs.NG-RES.2022.2","DOIUrl":null,"url":null,"abstract":"Embedded system applications usually have to meet real-time, energy or safety requirements on programs typically concurrently executed on a given MPSoC target platform. Enforcing such properties, e.g., by adapting the number of processors allocated to a program or by scaling the voltage/frequency mode of involved processors, is a difficult problem to solve, especially with a typically large varying environmental input (workload) per execution. In a previous work [4], we formalized the related enforcement problem using (a) finite state machines to model enforcement strategies, (b) discrete-time Markov chains to model the uncertain environment determining the system’s workload, and (c) the system response that defines the feedback for the reactive enforcer. In this paper, we apply that approach to specify and verify multi-requirement enforcement strategies and assess a case study for enforcing two independent requirements at the same time, i.e., latency and energy consumption. We evaluate and compare different enforcement strategies using probabilistic verification for the use case of an object detection application.","PeriodicalId":151755,"journal":{"name":"NG-RES@HiPEAC","volume":"6 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Multi-Requirement Enforcement of Non-Functional Properties on MPSoCs Using Enforcement FSMs - A Case Study\",\"authors\":\"Khalil Esper, S. Wildermann, J. Teich\",\"doi\":\"10.4230/OASIcs.NG-RES.2022.2\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Embedded system applications usually have to meet real-time, energy or safety requirements on programs typically concurrently executed on a given MPSoC target platform. Enforcing such properties, e.g., by adapting the number of processors allocated to a program or by scaling the voltage/frequency mode of involved processors, is a difficult problem to solve, especially with a typically large varying environmental input (workload) per execution. In a previous work [4], we formalized the related enforcement problem using (a) finite state machines to model enforcement strategies, (b) discrete-time Markov chains to model the uncertain environment determining the system’s workload, and (c) the system response that defines the feedback for the reactive enforcer. In this paper, we apply that approach to specify and verify multi-requirement enforcement strategies and assess a case study for enforcing two independent requirements at the same time, i.e., latency and energy consumption. We evaluate and compare different enforcement strategies using probabilistic verification for the use case of an object detection application.\",\"PeriodicalId\":151755,\"journal\":{\"name\":\"NG-RES@HiPEAC\",\"volume\":\"6 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1900-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"NG-RES@HiPEAC\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.4230/OASIcs.NG-RES.2022.2\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"NG-RES@HiPEAC","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4230/OASIcs.NG-RES.2022.2","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

摘要

嵌入式系统应用通常必须满足在给定MPSoC目标平台上并发执行的程序的实时性、能源或安全性要求。例如,通过调整分配给程序的处理器数量或通过缩放所涉及处理器的电压/频率模式来强制执行这些属性是一个很难解决的问题,特别是每次执行时通常有很大的变化环境输入(工作负载)。在之前的工作[4]中,我们使用(a)有限状态机来建模执行策略,(b)离散时间马尔可夫链来建模决定系统工作负载的不确定环境,以及(c)系统响应来定义被动执行器的反馈,从而形式化了相关的执行问题。在本文中,我们应用该方法来指定和验证多需求执行策略,并评估同时执行两个独立需求(即延迟和能耗)的案例研究。我们使用对象检测应用程序用例的概率验证来评估和比较不同的强制策略。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Multi-Requirement Enforcement of Non-Functional Properties on MPSoCs Using Enforcement FSMs - A Case Study
Embedded system applications usually have to meet real-time, energy or safety requirements on programs typically concurrently executed on a given MPSoC target platform. Enforcing such properties, e.g., by adapting the number of processors allocated to a program or by scaling the voltage/frequency mode of involved processors, is a difficult problem to solve, especially with a typically large varying environmental input (workload) per execution. In a previous work [4], we formalized the related enforcement problem using (a) finite state machines to model enforcement strategies, (b) discrete-time Markov chains to model the uncertain environment determining the system’s workload, and (c) the system response that defines the feedback for the reactive enforcer. In this paper, we apply that approach to specify and verify multi-requirement enforcement strategies and assess a case study for enforcing two independent requirements at the same time, i.e., latency and energy consumption. We evaluate and compare different enforcement strategies using probabilistic verification for the use case of an object detection application.
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
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:604180095
Book学术官方微信