The fast increasing presence of Internet-of-Things and fog computing resources exposes new challenges due to heterogeneity and non-negligible network delays among resources as well as the dynamism of operating conditions. Such a variable computing environment leads the applications to adopt an elastic and decentralized execution. To simplify the application deployment and run-time management, containers are widely used nowadays. The deployment of a container-based application over a geo-distributed computing infrastructure is a key task that has a significant impact on the application non-functional requirements (e.g., performance, security, cost). In this survey, we first develop a taxonomy based on the goals, the scope, the actions, and the methodologies considered to adapt at run-time the application deployment. Then, we use it to classify some of the existing research results. Finally, we identify some open challenges that arise for the application deployment in the fog. In literature, we can find many different approaches for adapting the containers deployment, each tailored for optimizing a specific objective, such as the application response time, its deployment cost, or the efficient utilization of the available computing resources. However, although several solutions for deploying containers exist, those explicitly considering the distinctive features of fog computing are at the early stages: indeed, existing solutions scale containers without considering their placement, or do not consider the heterogeneity, the geographic distribution, and mobility of fog resources.

Cardellini, V., Lo Presti, F., Nardelli, M., Rossi, F. (2020). Self-adaptive Container Deployment in the Fog: A Survey. In ALGOCLOUD 2019: Algorithmic Aspects of Cloud Computing (pp.77-102). Cham : Springer International Publishing [10.1007/978-3-030-58628-7_6].

Self-adaptive Container Deployment in the Fog: A Survey

Cardellini, Valeria;Lo Presti, Francesco;
2020-08-01

Abstract

The fast increasing presence of Internet-of-Things and fog computing resources exposes new challenges due to heterogeneity and non-negligible network delays among resources as well as the dynamism of operating conditions. Such a variable computing environment leads the applications to adopt an elastic and decentralized execution. To simplify the application deployment and run-time management, containers are widely used nowadays. The deployment of a container-based application over a geo-distributed computing infrastructure is a key task that has a significant impact on the application non-functional requirements (e.g., performance, security, cost). In this survey, we first develop a taxonomy based on the goals, the scope, the actions, and the methodologies considered to adapt at run-time the application deployment. Then, we use it to classify some of the existing research results. Finally, we identify some open challenges that arise for the application deployment in the fog. In literature, we can find many different approaches for adapting the containers deployment, each tailored for optimizing a specific objective, such as the application response time, its deployment cost, or the efficient utilization of the available computing resources. However, although several solutions for deploying containers exist, those explicitly considering the distinctive features of fog computing are at the early stages: indeed, existing solutions scale containers without considering their placement, or do not consider the heterogeneity, the geographic distribution, and mobility of fog resources.
5th International Symposium on Algorithmic Aspects of Cloud Computing (ALGOCLOUD 2019)
Munich, Germany
2019
Rilevanza internazionale
contributo
10-set-2019
ago-2020
Settore ING-INF/05 - SISTEMI DI ELABORAZIONE DELLE INFORMAZIONI
English
https://link.springer.com/chapter/10.1007/978-3-030-58628-7_6
Intervento a convegno
Cardellini, V., Lo Presti, F., Nardelli, M., Rossi, F. (2020). Self-adaptive Container Deployment in the Fog: A Survey. In ALGOCLOUD 2019: Algorithmic Aspects of Cloud Computing (pp.77-102). Cham : Springer International Publishing [10.1007/978-3-030-58628-7_6].
Cardellini, V; Lo Presti, F; Nardelli, M; Rossi, F
File in questo prodotto:
File Dimensione Formato  
algocloud2019.pdf

solo utenti autorizzati

Tipologia: Versione Editoriale (PDF)
Licenza: Copyright dell'editore
Dimensione 302.12 kB
Formato Adobe PDF
302.12 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS 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: https://hdl.handle.net/2108/253676
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 9
  • ???jsp.display-item.citation.isi??? 4
social impact