K. Nakauchi, F. Bronzino, Y. Shoji, I. Seskar, D. Raychaudhuri
{"title":"vMCN: virtual mobile cloud network for realizing scalable, real-time cyber physical systems","authors":"K. Nakauchi, F. Bronzino, Y. Shoji, I. Seskar, D. Raychaudhuri","doi":"10.1145/2955193.2955201","DOIUrl":null,"url":null,"abstract":"This paper presents virtual Mobile Cloud Network (vMCN), an architecture for scalable, real-time Cyber Physical Systems (CPS) based on virtualization-capable network infrastructure and highly distributed edge clouds. Emerging CPS applications running on mobile devices, such as Augmented Reality (AR) based navigation and self-driving cars, have fundamental limitations; (1) the response time over the networks is unmanageable and CPS applications suffer from large response times, especially over shared wireless links; (2) the number of real-virtual object mappings cannot be scaled to the approaching trillion order of magnitude in an unified manner. vMCN addresses these issues by introducing novel network virtualization techniques that exploit the \"named object\" abstraction provided by a fast and scalable global name service. Coordinating virtualized resources across multiple domains, vMCN supports the distributed edge cloud model by deploying an application aware anycast services that achieve the strict requirements of CPS while still scaling to the expected order of magnitude of devices. An initial vMCN prototype system has been developed using the ORBIT testbed resources and the NICT's virtualization-capable WiFi base station. Experimental results reveal the vMCN can support up to about 94% CPS cycles under the set goal of 100 ms, outperforming the baseline system by almost two times.","PeriodicalId":91161,"journal":{"name":"Proceedings. Data Compression Conference","volume":"80 1","pages":"2:1-2:6"},"PeriodicalIF":0.0000,"publicationDate":"2016-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings. Data Compression Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2955193.2955201","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 9
Abstract
This paper presents virtual Mobile Cloud Network (vMCN), an architecture for scalable, real-time Cyber Physical Systems (CPS) based on virtualization-capable network infrastructure and highly distributed edge clouds. Emerging CPS applications running on mobile devices, such as Augmented Reality (AR) based navigation and self-driving cars, have fundamental limitations; (1) the response time over the networks is unmanageable and CPS applications suffer from large response times, especially over shared wireless links; (2) the number of real-virtual object mappings cannot be scaled to the approaching trillion order of magnitude in an unified manner. vMCN addresses these issues by introducing novel network virtualization techniques that exploit the "named object" abstraction provided by a fast and scalable global name service. Coordinating virtualized resources across multiple domains, vMCN supports the distributed edge cloud model by deploying an application aware anycast services that achieve the strict requirements of CPS while still scaling to the expected order of magnitude of devices. An initial vMCN prototype system has been developed using the ORBIT testbed resources and the NICT's virtualization-capable WiFi base station. Experimental results reveal the vMCN can support up to about 94% CPS cycles under the set goal of 100 ms, outperforming the baseline system by almost two times.