With the innovation of computer technology and the rapid development of the Internet, cloud computing came into being. In cloud computing, resources and tasks are dynamic and heterogeneous. In this thesis, after studying the task scheduling problem in cloud computing and analyzing the existing problems of resource allocation algorithm, we found that cloud computing usually needs to deal with a large number of computing tasks. Therefore, how to allocate computing resources reasonably and efficiently to schedule tasks, so that the tasks can be completed in a shorter time and at a lower cost, is an important issue. During this studying, a task scheduling algorithm based on genetic algorithm is proposed and implemented. The results generated by this algorithm can make the task completion time shorter and at the same time also cost less. We also introduce time and cost factors into the fitness function to control the evolution direction of the population so that we can meet the different needs and preferences of users. Finally, the simulation of the above proposed algorithm is executed by expanding the cloud computing simulation platform, CloudSim. Then we analyze its performance through experiments. Experimental results show that our algorithm is an effective task scheduling algorithm in cloud computing.
Con l'innovazione della tecnologia informatica e il rapido sviluppo di Internet, è nato il cloud computing. Nel cloud computing, risorse e attività sono dinamiche ed eterogenee. In questa tesi, dopo aver studiato il problema della pianificazione delle attività nel cloud computing e dopo aver analizzato i problemi esistenti riguardanti gli algoritmi di allocazione delle risorse, abbiamo scoperto che il cloud computing, in genere, ha bisogno di gestire un gran numero di attività di elaborazione. Pertanto, come allocare le risorse di elaborazione in modo ragionevole ed efficiente per pianificare le attività, in modo che tutte le attività possano essere completate in un tempo più breve e ad un costo inferiore, è una questione importante. Durante questo studio, viene proposto e implementato un algoritmo di pianificazione delle attività basato sugli algoritmi genetici. I risultati generati da questo algoritmo possono ridurre il tempo di completamento dell'attività e allo stesso tempo anche costare meno. Introduciamo anche fattori di tempo e costo nella funzione di fitness per controllare la direzione evolutiva della popolazione in modo da soddisfare le diverse esigenze e preferenze degli utenti. Infine, la simulazione dell'algoritmo sopra proposto viene eseguita espandendo la piattaforma di simulazione del cloud computing, CloudSim. Poi analizziamo le sue prestazioni attraverso esperimenti. I risultati ottenuti mosstrano che l'algoritmo di pianificazione delle attività risulta efficace nel cloud computing.
Optimization of task scheduling in heterogeneous Cloud environment using genetic algorithm
ZHANG, LIDONG
2017/2018
Abstract
With the innovation of computer technology and the rapid development of the Internet, cloud computing came into being. In cloud computing, resources and tasks are dynamic and heterogeneous. In this thesis, after studying the task scheduling problem in cloud computing and analyzing the existing problems of resource allocation algorithm, we found that cloud computing usually needs to deal with a large number of computing tasks. Therefore, how to allocate computing resources reasonably and efficiently to schedule tasks, so that the tasks can be completed in a shorter time and at a lower cost, is an important issue. During this studying, a task scheduling algorithm based on genetic algorithm is proposed and implemented. The results generated by this algorithm can make the task completion time shorter and at the same time also cost less. We also introduce time and cost factors into the fitness function to control the evolution direction of the population so that we can meet the different needs and preferences of users. Finally, the simulation of the above proposed algorithm is executed by expanding the cloud computing simulation platform, CloudSim. Then we analyze its performance through experiments. Experimental results show that our algorithm is an effective task scheduling algorithm in cloud computing.| File | Dimensione | Formato | |
|---|---|---|---|
|
Optimization task scheduling in heterogeneous cloud environment using GA.pdf
Open Access dal 05/04/2020
Descrizione: Thesis PDF
Dimensione
1.21 MB
Formato
Adobe PDF
|
1.21 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/10589/147421