umu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Location-aware load prediction in edge data centers
Umeå University, Faculty of Science and Technology, Department of Computing Science. (Distributed Systems)ORCID iD: 0000-0002-9156-3364
Umeå University, Faculty of Science and Technology, Department of Computing Science. (Distributed Systems)
Umeå University, Faculty of Science and Technology, Department of Computing Science. (Distributed Systems)
2017 (English)In: 2017 Second International Conference on Fog and Mobile Edge Computing (FMEC), IEEE, 2017, 25-31 p.Conference paper, Published paper (Other academic)
Abstract [en]

Mobile Edge Cloud (MEC) is a platform complementing traditional centralized clouds, consisting in moving computing and storage capacity closer to users -e. g., as Edge Data Centers (EDC) in base stations -in order to reduce application-level latency and network bandwidth. The bounded coverage radius of base station and limited capacity of each EDC intertwined with user mobility challenge the operator's ability to perform capacity adjustment and planning. To face this challenge, proactive resource provisioning can be performed. The resource usage in each EDC is estimated in advance, which is made available for the decision making to efficiently determine various management actions and ensure that EDCs persistently satisfies the Quality of Service (QoS), while maximizing resource utilization. In this paper, we propose location-aware load prediction. For each EDC, load is not only predicted using its own historical load time series -as done for centralized clouds -but also those of its neighbor EDCs. We employ Vector Autoregression Model (VAR) in which the correlation among adjacent EDCs load time series are exploited. We evaluate our approach using real world mobility traces to simulate load in each EDC and conduct various experiments to evaluate the proposed algorithm. Result shows that our proposed algorithm is able to achieve an average accuracy of up to 93% on EDCs with substantial average load, which slightly improves prediction by 4.3% compared to the state-of-the-art approach. Considering the expected scale of MEC, this translates to substantial cost savings e. g., servers can be shutdown without QoS violation.

Place, publisher, year, edition, pages
IEEE, 2017. 25-31 p.
Keyword [en]
Workload Prediction, Proactive Resource Management, Mobile Edge Cloud, VAR Model, User Mobility
National Category
Computer Systems
Identifiers
URN: urn:nbn:se:umu:diva-135700ISI: 000411731700004ISBN: 978-1-5386-2859-1 (print)OAI: oai:DiVA.org:umu-135700DiVA: diva2:1105213
Conference
The 2nd International Conference on Fog and Mobile Edge Computing (FMEC), May 8-11, 2017, Valencia, Spain
Projects
WASP
Available from: 2017-06-02 Created: 2017-06-02 Last updated: 2017-10-16Bibliographically approved

Open Access in DiVA

No full text

Search in DiVA

By author/editor
Nguyen, Chanh Le TanKlein, CristianElmroth, Erik
By organisation
Department of Computing Science
Computer Systems

Search outside of DiVA

GoogleGoogle Scholar

Total: 124 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf