摘要:多核架构下多临界实时应用的I/O争用感知映射

Laure Abdallah, M. Jan, Jérôme Ermont, C. Fraboul
{"title":"摘要:多核架构下多临界实时应用的I/O争用感知映射","authors":"Laure Abdallah, M. Jan, Jérôme Ermont, C. Fraboul","doi":"10.1109/RTAS.2016.7461348","DOIUrl":null,"url":null,"abstract":"Many-core architectures are more promising hardware to design real-time systems than multi-core systems as they should enable an easier mastered integration of an higher number of applications, potentially of different level of criticalities. However, the worst-case behavior of the Network-on-Chip (NoC) for both inter-core and core-to-Input/Output (I/O) communications of critical applications must be established. We use the term core-to-I/O for both core communications from or to I/O interfaces. The mapping over the NoC of both critical and non-critical applications has an impact on the network contention these critical communications exhibit. So far, all existing mapping strategies have focused on inter-core communications. However, we claim that many-cores in embedded real-time systems will be integrated within backbone ethernet networks, as they mostly provide ethernet controllers as I/O interfaces. In this work, we first show that ethernet packets can be dropped due to an internal congestion in the NoC, if these core-to-I/O communications are not taken into account while mapping applications. To this end, we rely on a case study from the avionic domain. It is made of a critical Full Authority Digital Engine (FADEC) application and a non-critical Health Monitoring (HM) application of the engine, used for recognizing incipient failure conditions. Based on this analysis, we introduce our approach to map critical and non critical real-time applications over many-cores that reduces the WCTT of core-to-I/O communications. We show for two variants of our case study that our algorithm successfully find a mapping that avoids ethernet packets, whose payload are making the core-to-I/O communications, to be dropped. This demonstrates the benefits of our proposal compared to a state of the art mapping strategy that fails to do so.","PeriodicalId":338179,"journal":{"name":"2016 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Poster Abstract: I/O Contention Aware Mapping of Multi-Criticalities Real-Time Applications over Many-Core Architectures\",\"authors\":\"Laure Abdallah, M. Jan, Jérôme Ermont, C. Fraboul\",\"doi\":\"10.1109/RTAS.2016.7461348\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Many-core architectures are more promising hardware to design real-time systems than multi-core systems as they should enable an easier mastered integration of an higher number of applications, potentially of different level of criticalities. However, the worst-case behavior of the Network-on-Chip (NoC) for both inter-core and core-to-Input/Output (I/O) communications of critical applications must be established. We use the term core-to-I/O for both core communications from or to I/O interfaces. The mapping over the NoC of both critical and non-critical applications has an impact on the network contention these critical communications exhibit. So far, all existing mapping strategies have focused on inter-core communications. However, we claim that many-cores in embedded real-time systems will be integrated within backbone ethernet networks, as they mostly provide ethernet controllers as I/O interfaces. In this work, we first show that ethernet packets can be dropped due to an internal congestion in the NoC, if these core-to-I/O communications are not taken into account while mapping applications. To this end, we rely on a case study from the avionic domain. It is made of a critical Full Authority Digital Engine (FADEC) application and a non-critical Health Monitoring (HM) application of the engine, used for recognizing incipient failure conditions. Based on this analysis, we introduce our approach to map critical and non critical real-time applications over many-cores that reduces the WCTT of core-to-I/O communications. We show for two variants of our case study that our algorithm successfully find a mapping that avoids ethernet packets, whose payload are making the core-to-I/O communications, to be dropped. This demonstrates the benefits of our proposal compared to a state of the art mapping strategy that fails to do so.\",\"PeriodicalId\":338179,\"journal\":{\"name\":\"2016 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS)\",\"volume\":\"28 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-04-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/RTAS.2016.7461348\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/RTAS.2016.7461348","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

摘要

多核体系结构是比多核系统更有希望设计实时系统的硬件,因为它们应该能够更容易地集成更多的应用程序,这些应用程序可能具有不同的临界级别。然而,对于关键应用程序的核间和核到输入/输出(I/O)通信,必须确定片上网络(NoC)的最坏情况行为。我们使用术语核心到I/O来表示来自I/O接口或到I/O接口的核心通信。关键和非关键应用程序在NoC上的映射对这些关键通信所显示的网络争用有影响。到目前为止,所有现有的映射策略都侧重于核间通信。然而,我们声称嵌入式实时系统中的多核将集成在骨干以太网网络中,因为它们主要提供以太网控制器作为I/O接口。在这项工作中,我们首先表明,如果在映射应用程序时不考虑这些核心到i /O通信,则由于NoC中的内部拥塞,以太网数据包可能会被丢弃。为此,我们依靠航空电子领域的一个案例研究。它由关键的完全授权数字引擎(FADEC)应用程序和引擎的非关键运行状况监控(HM)应用程序组成,用于识别早期故障条件。基于此分析,我们介绍了在多核上映射关键和非关键实时应用程序的方法,从而降低了核心到i /O通信的WCTT。对于案例研究的两个变体,我们展示了我们的算法成功地找到了一个映射,该映射避免了以太网数据包(其有效负载正在进行核心到i /O通信)被丢弃。这证明了我们的建议与不能这样做的最先进的映射策略相比的好处。
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
Poster Abstract: I/O Contention Aware Mapping of Multi-Criticalities Real-Time Applications over Many-Core Architectures
Many-core architectures are more promising hardware to design real-time systems than multi-core systems as they should enable an easier mastered integration of an higher number of applications, potentially of different level of criticalities. However, the worst-case behavior of the Network-on-Chip (NoC) for both inter-core and core-to-Input/Output (I/O) communications of critical applications must be established. We use the term core-to-I/O for both core communications from or to I/O interfaces. The mapping over the NoC of both critical and non-critical applications has an impact on the network contention these critical communications exhibit. So far, all existing mapping strategies have focused on inter-core communications. However, we claim that many-cores in embedded real-time systems will be integrated within backbone ethernet networks, as they mostly provide ethernet controllers as I/O interfaces. In this work, we first show that ethernet packets can be dropped due to an internal congestion in the NoC, if these core-to-I/O communications are not taken into account while mapping applications. To this end, we rely on a case study from the avionic domain. It is made of a critical Full Authority Digital Engine (FADEC) application and a non-critical Health Monitoring (HM) application of the engine, used for recognizing incipient failure conditions. Based on this analysis, we introduce our approach to map critical and non critical real-time applications over many-cores that reduces the WCTT of core-to-I/O communications. We show for two variants of our case study that our algorithm successfully find a mapping that avoids ethernet packets, whose payload are making the core-to-I/O communications, to be dropped. This demonstrates the benefits of our proposal compared to a state of the art mapping strategy that fails to do so.
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
期刊最新文献
Trading Cores for Memory Bandwidth in Real-Time Systems A Kernel for Energy-Neutral Real-Time Systems with Mixed Criticalities Poster Abstract: Scheduling Multi-Threaded Tasks to Reduce Intra-Task Cache Contention Demo Abstract: Predictable SoC Architecture Based on COTS Multi-Core TaskShuffler: A Schedule Randomization Protocol for Obfuscation against Timing Inference Attacks in Real-Time Systems
×
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