Authors
Anang Hudaya Muhamad Amin and Asad Khan (Monash University)
Abstract
Cloud computing is a latest trend in distributed computing paradigm. Data management within cloud involves extreme parallelisation and distribution of data. In this context, storage and retrieval processes within cloud systems are becoming more complex and require an efficient data management mechanism, especially in the activities involving real-time data processing. Current data management scheme should be able to manage data partitioning between processing nodes, as well as the capability to collaborate nodes for a specific task efficiently. Currently, these two features are still lacking in the available schemes. In this poster, we present our current works on distributed data management scheme for cloud computing that enables data access to be conducted effectively by the means of an associative computing approach. We intend to treat data as pattern and perform data storage and retrieval through a supervised pattern recognition mechanism. This approach envisages data retrieval to be implemented as a distributed pattern association process that is implemented through the integration of loosely-coupled computational networks. Followed by a divide-and-distribute approach that allows distribution of these networks within the cloud dynamically. The proposed scheme enables scalable mechanism for data retrieval and storage within cloud system.