A proposed novel enterprise cloud development application model

Currently, there is limited guidance on generic cloud software architecture model for designing and building cloud applications and each platform provider has different standards that influences the way applications are developed and written. In this paper we propose a cloud application development model based on developing services (typically REST API services). We see clear benefits in applying this model to service oriented architecture (SOA) and business process management (BPM) design in order to develop enterprise applications for cloud environments in a systematic, rapid manner. This conceptual model described in this paper is drawn from the core ideas and practices behind REST API, BPM, and SOA. In this paper we focus on the proposed software development model for enterprise cloud applications and services. Throughout this paper we use the term cloud application and cloud service interchangeably. The model described in this paper can be applied to both. Further research on a software-modeling tool that provides a standardized structure for cloud software development and deployment seems to be in order. Supported by a graphical modeling environment, such a tool would utilize a Cloud Resource Provisioning language, will be able to seamlessly deploy and provision cloud resources across multiple cloud service providers. The presented research provides a practical and extensible application development model for cloud platforms. The model described can be used in a variety of applications as well as migration of legacy code to cloud platforms. By using this model, a cloud application can be deployed as small functionally separate components to different cloud nodes. Resource allocation can be defined for these nodes on a per need basis, which minimizes the cost of acquiring a fail over and load balanced system compared to the traditional on premise application development. In this paper, a detailed description of the analysis and development steps of the proposed model is also presented.