Enormous amounts of data are generated from different types of devices, includ- ing IOT devices, embedded sensors and personal computers. How to effectively manage these data without breaking related constraints and privacy policies has become a crucial and valuable issue for data-intensive applications, especially in the Fog Computing environment where both resources at the edge of the network and resources in the cloud are involved. Fog Computing extends the capability of Cloud Computing by exploiting the potential of the edge of the network. With Fog Computing, we can take ad- vantage of both resources on the edge (less latency) and on the cloud (scalabil- ity,reliability). There are two possibility when it comes to manage huge volume of data between cloud and edge of the network: to move the data or to move the computation tasks. In this thesis, we mainly focus on the computation task movement. To enable computation task movement, a DaaS approach applied to a Fog environment has been adopted. Our goal is to propose a user-friendly DaaS platform for the data-intensive application developers where data processing occurs not only on cloud but also at the edge of the network, taking into account the heterogeneous IoT devices, network delay and privacy issues. The proposed platform, which is designed under the principles of Service Oriented Computing, is able to provide the monitoring ability and the movement decisions.
Enormi quantit`a di dati vengono generate da diversi tipi di dispositivi, inclusi dispositivi IOT, sensori incorporati e personal computer. Come gestire in modo efficace questi dati senza infrangere i vincoli e le politiche sulla privacy `e diven- tato un problema cruciale e prezioso per le applicazioni ad alta intensit`a di dati, specialmente nell’ambiente di Fog Computing in cui sono coinvolte sia le risorse ai margini della rete sia le risorse nel cloud. Fog Computing estende la capacit`a del Cloud Computing sfruttando il poten- ziale del bordo della rete. Con Fog Computing, possiamo sfruttare entrambe le risorse sul bordo (meno latenza) e sul cloud (scalabilit`a, affidabilit`a). Ci sono due possibilit`a quando si tratta di gestire un volume enorme di dati tra il cloud e il bordo della rete: spostare i dati o spostare le attivit`a di calcolo. In questa tesi, ci concentriamo principalmente sul movimento del compito di calcolo. Per consentire il movimento delle attivit`a di calcolo, `e stato adottato un approccio DaaS applicato a un ambiente di nebbia. Il nostro obiettivo `e proporre una piattaforma DaaS user-friendly per gli sviluppatori di applicazioni ad alta intensit`a di dati in cui l’elaborazione dei dati avviene non solo su cloud ma anche ai margini della rete, tenendo conto di disposi- tivi IoT eterogenei, ritardo di rete e problemi di privacy. La piattaforma proposta, progettata secondo i principi di Service Oriented Computing, `e in grado di fornire le capacit`a di monitoraggio e le decisioni relative ai movimenti.
Improving DaaS with Kubernetes in fog computing environments
SUN, CHAO
2017/2018
Abstract
Enormous amounts of data are generated from different types of devices, includ- ing IOT devices, embedded sensors and personal computers. How to effectively manage these data without breaking related constraints and privacy policies has become a crucial and valuable issue for data-intensive applications, especially in the Fog Computing environment where both resources at the edge of the network and resources in the cloud are involved. Fog Computing extends the capability of Cloud Computing by exploiting the potential of the edge of the network. With Fog Computing, we can take ad- vantage of both resources on the edge (less latency) and on the cloud (scalabil- ity,reliability). There are two possibility when it comes to manage huge volume of data between cloud and edge of the network: to move the data or to move the computation tasks. In this thesis, we mainly focus on the computation task movement. To enable computation task movement, a DaaS approach applied to a Fog environment has been adopted. Our goal is to propose a user-friendly DaaS platform for the data-intensive application developers where data processing occurs not only on cloud but also at the edge of the network, taking into account the heterogeneous IoT devices, network delay and privacy issues. The proposed platform, which is designed under the principles of Service Oriented Computing, is able to provide the monitoring ability and the movement decisions.File | Dimensione | Formato | |
---|---|---|---|
thesis-sun-chao.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Thesis paper
Dimensione
1.64 MB
Formato
Adobe PDF
|
1.64 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/140107