Adaptive clustering of Glassfish application server in Amazon EC2 cloud

OData support
Imre Gábor
Department of Automation and Applied Informatics

My thesis is about a cluster management application written in Java language operated in a cloud. The software manages the cluster resources running in the cloud with the help of procedures called from the program. Its task is to enable the optimal utilization of the resources by the Glassfish application server operated in the cloud.

My program substitutes the solutions of Amazon service management with this task being performed by the application I prepared. To solve the problem, I used a statistical formula to establish the resource requirements matching the number of requests varying with time.

A further objective was not to have my application running on the client side, but in the infrastructure of the Amazon cloud infrastructure instead, in a single environment with the cluster. Beside the above, the current operational status of my application being viewable with the help of a client side program was another goal.

The performance test showed that the Glassfish cluster is able to operate in the system of the Amazon cloud and utilizes the necessary resources according to the algorithm applied.


Please sign in to download the files of this thesis.