Service-oriented collaborative computing management using kepler extension model

Abstract

Collaborative computing supports a cooperative work with computers and networks among people. By adapting workflow in collaborative computing, it is easier to manage complex collaboration since it provides automation of works. Furthermore, collaborators can understand what activities of the other collaborators are and collaborative computing makes easy how to coordinate each work. To support more complex and dynamic activities, especially human interaction and on demand work reconfiguration, during collaboration, it is necessary to reconfigure workflow at run time. In this thesis, we focus on two issues, which are supporting of scalability within a limited cost boundary, and workflow configuration management according to Service Level Agreement (SLA) changes under collaborative computing environment. To discuss issues for collaborative computing, we propose a service-oriented computing to satisfy the service level based on Grid workflow model. The main contributions are configuration management when service level is given for a specific collaboration computing with kepler workflow model, integration module for workflow management and scheduling, and a prototype implementation of e-Organ Simulator Integrated Collaborative Computing (OSICC) application, respectively. Finally, to evaluate our proposed system, we examined experiments with a new drug discovery application to show the characteristics and advantages of our system compare with performance comparison. The result shows that our proposed system can flexibly enhance the existing workflow management in terms of SLA guarantee that user SLA changes dynamically. In aspect of performance and functionality, the proposed system can satisfy collaborators in different domains.