Diploma and Master Theses (authored and supervised):
"Towards Providing Unified Access To Cloud Data Storage Services";
Supervisor: S. Dustdar, Ph. Leitner;
Institut für Informationssysteme, AB Verteilte Systeme,
final examination: 2013-11-18.
With an increasing number of devices using applications, the amount of data that has to be managed has grown rapidly. This means, that scalability of databases gets more and more important. So in the last years so called not only SQL (NoSQL) databases have become popular
because they can be distributed more easy than traditional relational databases.The problem is, that different data stores use different ways to manage their data and provide specific interfaces
to persist and load this data sets. This makes it hard for application developers to use this new technologies in their applications. So using different types of persistence, as described in the polyglot persistence model, means that a lot of code has to be written for covering the different APIs.
The following thesis describes and presents a solution to access different types of data storage through one interface. Therefore, the differences between the management and storage of data are described and discussed. Based on this discussion, a common interface will be designed. Because the API can not cover all different unctionalities, developers will also get the possibility to access specific libraries, to use special functions for a data store. This means, that data can be stored where it fits best, without losing any functionality or flexibility. In some cases, it is useful
to move data from one data store to another. This often requires a lot of migration code to read and write data. In this thesis, we will also present a solution for this problem and a prototype implementation of this feature will be discussed.
Created from the Publication Database of the Vienna University of Technology.