In cloud computing, elasticity is defined as "the degree to which a system is able to adapt to workload changes by provisioning and de-provisioning resources in an autonomous manner, such that at each point in time the available resources match the current demand as closely as possible". Elasticity term, inherited from physics and economics fields, is nowadays heavily used in cloud computing. In this context, elasticity is commonly understood as the ability of a system to automatically provision and de-provision computing resources on demand, as workloads change. However, it must be noted that the definition of elasticity is largely inconsistent and unspecific. In fact, comparing methodologies and systems often leads to confusion and it’s required to consider also efficiency and scalability aspects in order to have a representative metric to compare different systems. Initially, diverse related works will be reviewed, in order to define the state of the art of elasticity in the cloud. In a second moment, a number of elasticity mechanisms will be analyzed, based on the main features found in several commercial and academic solutions. A discussion about some of the challenges and open issues associated with the use of elasticity features in cloud computing will follow. In order to show how the elasticity works in practice, a static analyser has been used as a starting point to apply some of the techniques presented. The project developed in this work extended an open source cloud simulator, namely CloudSim Plus, for the purpose of demonstrating the techniques validity, comparing the results obtained processing a workload trace. The objective of this paper is to show, through simulation, the effectiveness of cloud resources adaptation techniques.
Nell’ambito del cloud computing, l’elasticità (elasticity) è definita come il “grado in cui un sistema è capace di adattarsi a cambiamenti di carico, grazie al provisioning e al de-provisioning di risorse in modo automatico, in modo che in qualsiasi momento le risorse disponibili uguaglino il più possibile la domanda esistente. Tuttavia, si noti che la definizione di elasticità è in gran parte incoerente e non specifica. Infatti, il confronto di metodologie e sistemi differenti spesso causa confusione. È necessario considerare anche gli aspetti di efficienza e scalabilità per avere una metrica rappresentativa nel confronto di sistemi differenti. Inizialmente, saranno esaminati diversi lavori correlati, al fine di definire lo stato dell'arte dell'elasticità nel cloud computing. In un secondo momento, saranno analizzati alcuni meccanismi di elasticità, basati sulle caratteristiche principali presenti in alcune soluzioni commerciali e accademiche. Seguirà una discussione su alcune delle sfide e dei problemi aperti associati all'uso delle funzionalità di elasticità nel cloud computing. Per mostrare come funziona l'elasticità in pratica, uno static analyser è stato utilizzato come punto di partenza per applicare alcune delle tecniche presentate. Il progetto sviluppato in questo lavoro ha esteso un simulatore cloud open source, ovvero CloudSim Plus, allo scopo di dimostrare la validità di tali tecniche, confrontando i risultati ottenuti nell'elaborazione di una traccia di workload. L'obiettivo di questo lavoro è di dimostrare, attraverso alcune simulazioni, l'efficacia delle tecniche di adattamento delle risorse cloud.
Techniques for the management of cloud resources adaptations applied on CloudSim Plus simulator
ZANGARI, LUCA
2017/2018
Abstract
In cloud computing, elasticity is defined as "the degree to which a system is able to adapt to workload changes by provisioning and de-provisioning resources in an autonomous manner, such that at each point in time the available resources match the current demand as closely as possible". Elasticity term, inherited from physics and economics fields, is nowadays heavily used in cloud computing. In this context, elasticity is commonly understood as the ability of a system to automatically provision and de-provision computing resources on demand, as workloads change. However, it must be noted that the definition of elasticity is largely inconsistent and unspecific. In fact, comparing methodologies and systems often leads to confusion and it’s required to consider also efficiency and scalability aspects in order to have a representative metric to compare different systems. Initially, diverse related works will be reviewed, in order to define the state of the art of elasticity in the cloud. In a second moment, a number of elasticity mechanisms will be analyzed, based on the main features found in several commercial and academic solutions. A discussion about some of the challenges and open issues associated with the use of elasticity features in cloud computing will follow. In order to show how the elasticity works in practice, a static analyser has been used as a starting point to apply some of the techniques presented. The project developed in this work extended an open source cloud simulator, namely CloudSim Plus, for the purpose of demonstrating the techniques validity, comparing the results obtained processing a workload trace. The objective of this paper is to show, through simulation, the effectiveness of cloud resources adaptation techniques.| File | Dimensione | Formato | |
|---|---|---|---|
|
Luca_Zangari_Thesis.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Luca Zangari Thesis
Dimensione
1.22 MB
Formato
Adobe PDF
|
1.22 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/10589/144783