Fog-Assisted Aggregated Synchronization Scheme for Mobile Cloud Storage Applications

Giwon Lee, Haneul Ko, Sangheon Pack, Valentino Pacifici, György Dán

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Cloud storage applications, such as Dropbox and Google Drive, have recently become very popular among mobile users. In these applications, a cloud server is responsible for synchronizing updates to files among mobile users, and thus if files are shared by many mobile users and are frequently updated then the resulting synchronization traffic can be significant. In order to reduce the synchronization traffic with providing acceptable access latency, we propose a fog-assisted aggregated synchronization (FAS) scheme in which the fog computing server and the cloud server conduct localized and aggregated synchronizations, respectively. We develop an analytical model of the FAS scheme based on renewal-reward theory and use it for model-based adjustment of the timer that controls the trade-off between access latency and synchronization traffic. We use analytical and simulation results to give insight into the effects of the timer, the update-to-access ratio, the number of mobile users, and the sensitivity to the arrival process. The analytical and simulation results demonstrate that the FAS scheme can reduce the synchronization traffic significantly with acceptable access latency compared to conventional schemes.

Original languageEnglish
Article number8704940
Pages (from-to)56852-56863
Number of pages12
JournalIEEE Access
Volume7
DOIs
Publication statusPublished - 2019 Jan 1

Keywords

  • fog-assisted aggregated synchronization
  • Mobile cloud storage
  • renewal-reward theory
  • synchronization traffic

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Fingerprint Dive into the research topics of 'Fog-Assisted Aggregated Synchronization Scheme for Mobile Cloud Storage Applications'. Together they form a unique fingerprint.

  • Cite this