SOURCE:

Faculty: Engineering
Department: Electronic And Computer Engineering

CONTRIBUTORS:

Nwazor, N. O.
Inyiama, H. C.

ABSTRACT:

This dissertation titled “Modelling a Cloud-based Collaboration Platform with Enhanced Security” is aimed at improving the process of research collaboration especially in Nigeria. The proposed system is a result of a search for a system that will serve as a solution to some of the following challenges encountered in engineering and technology research: research duplication, poor collaboration, poor funding/ wastage of the available funds, inadequate infrastructure, inadequate storage for the large volume of data generated in research, serious brain drain and poor awareness creation among the key stakeholders to mention but a few. The platform referred to as Cloud Research Collaboration Model (CRCM) which has the capability of automatic data acquisition from research equipments was designed based on Platform as a Service cloud computing models. It features: Content Management System (CMS), Document and File Sharing (DFS), Real Time Chatting (RTC), data acquisition from research equipments, online research library, research tracking and information management system. A modified form of waterfall methodology, discrete event modeling and formal method were used in different aspects of the work. The review of existing cloud based research collaboration platforms carried out showed that the two major challenges still facing cloud computing applications include task scheduling and the issue of data integrity and security in the cloud. Thus the need for an enhanced adaptive queuing model to improve task scheduling and an enhanced security system to help improve data security in the cloud since cloud computing involves handing one’s data over to a third party as proposed in this research work. The proposed security system, Hybrid Encryption Algorithm (HEA) which is a combination of a form of advanced encryption standards algorithm and digital signature will be implemented at the application layer before the data is sent to the cloud. The proposed queuing model which is an enhanced G1/G/1 queuing model was modelled using Matlab and simulated using SimEvent tool. Simulations were carried out to validate the model and test its performance using Optical Network simulation tools using parameters from Cisco to create a virtual data center and it gave values similar to the results obtained from the SimEvent with a very low page response of 0.032seconds and server utilisation of 90% which is an improvement on the current queuing systems. Test deployment was done at Cisco systems in Lagos to further validate the system.