Mobile Cloud Computing is an emergent topic that has received increasing attention considering the importance and the practical use of mobile applications. A relevant research direction in this field aims at optimizing the offloading decisions of mobile applications based on different metrics, such as performance and energy consumption, and according to the dynamic environment conditions in which the application is located. In this paper we define and implement two parallel algorithms for supporting offloading decisions at runtime. They are based on the well-known Depth First Search and Dijkstra shortest path algorithms, properly customized for the mobile cloud domain. Considering the mobile characteristics and the application requirements, the algorithms permit to optimize the effectiveness of mobile applications providing an optimal offloading strategy to be followed during the application execution. The proposed algorithms are experimentally validated by comparing their performance and scalability, also with respect to a previously defined approach based on model checking. The comparison is supported by a simulator tool that we developed on purpose to assess the potentiality of the algorithms at runtime. The achieved results prove the applicability of the approach into practice, which is also witnessed by a case study focusing on a mobile cloud application for public security

Runtime  Computation of  Optimal Offloading Scheduling  / MORICHETTA, Andrea; Barbara Re; Francesco Tiezzi. - STAMPA. - (2018), pp. 73-78. ((Intervento presentato al convegno 6th IEEE International Conference on Mobile Cloud Computing, Services, and Engineering tenutosi a Bamberg nel March 26 – March 29, 2018 [10.1109/MobileCloud.2018.00019].

Runtime  Computation of  Optimal Offloading Scheduling 

Francesco Tiezzi
2018

Abstract

Mobile Cloud Computing is an emergent topic that has received increasing attention considering the importance and the practical use of mobile applications. A relevant research direction in this field aims at optimizing the offloading decisions of mobile applications based on different metrics, such as performance and energy consumption, and according to the dynamic environment conditions in which the application is located. In this paper we define and implement two parallel algorithms for supporting offloading decisions at runtime. They are based on the well-known Depth First Search and Dijkstra shortest path algorithms, properly customized for the mobile cloud domain. Considering the mobile characteristics and the application requirements, the algorithms permit to optimize the effectiveness of mobile applications providing an optimal offloading strategy to be followed during the application execution. The proposed algorithms are experimentally validated by comparing their performance and scalability, also with respect to a previously defined approach based on model checking. The comparison is supported by a simulator tool that we developed on purpose to assess the potentiality of the algorithms at runtime. The achieved results prove the applicability of the approach into practice, which is also witnessed by a case study focusing on a mobile cloud application for public security
Mobile Cloud Computing, Services, and Engineering
6th IEEE International Conference on Mobile Cloud Computing, Services, and Engineering
Bamberg
March 26 – March 29, 2018
MORICHETTA, Andrea; Barbara Re; Francesco Tiezzi
File in questo prodotto:
File Dimensione Formato  
main.pdf

Accesso chiuso

Dimensione 352.18 kB
Formato Adobe PDF
352.18 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/2158/1243608
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact