OAK

SmartX multi-view visibility framework with flow-centric visibility for SDN-enabled multisite cloud playground

Metadata Downloads
Abstract
Modern information communication technologies (ICT) infrastructures are getting complicated to cope with the various demands needed to accommodate the emerging technology paradigms such as cloud, software-defined networking (SDN), and internet of things (IoT). Visibility is essential for the effective operation of such modern ICT infrastructures to easily pinpoint server faults, network bottlenecks, and application performance troubles. Even though many conventional monitoring solutions are now available, multi-layer visibility is still limited when operating such complicated infrastructures. To address this particular limitation, a futuristicmulti-layer visibility framework denoted as SmartX multi-view visibility framework (MVF), is proposed for unifying the visibility of underlay, physical and virtual resources, flow, and workload layers. To unify multi-layer visibility, this paper presents a comprehensive extension of SmartX MVF with flow-centric visibility for simultaneously monitoring physical-virtual resources, flows classification, and visualization to eventually assist secured operation of SDN-enabled multisite cloud infrastructure. Flow-centric visibility design has five main components (1) a lightweight network packets-precise flows visibility collection component, (2) a visibility data aggregation and tagging component, (3) a multi-layer visibility data integration component, (4) a non-learning-based network packets flows classification component, and (5) a visualization component. Furthermore, a prototype implementation of SmartXMVF with flow-centric visibility is deployed in an SDN-enabled multisite cloud playground to verify the proposed multi-view visibility of fine-grained flow-aware physical-virtual resources. © 2019 by the authors.
Author(s)
Usman, MuhammadRathore, Muhammad AhmadKim, JongWon
Issued Date
2019-05
Type
Article
DOI
10.3390/app9102045
URI
https://scholar.gist.ac.kr/handle/local/12713
Publisher
MDPI AG
Citation
Applied Sciences (Switzerland), v.9, no.10
ISSN
2076-3417
Appears in Collections:
Department of AI Convergence > 1. Journal Articles
공개 및 라이선스
  • 공개 구분공개
파일 목록

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.