AuthorsS. Schork, F. Zahid, D. Pradhan, S. Kicin and A. Schwichtenberg
EditorsF. Bordeleau, A. Sillitti, P. Meirelles and V. Lenarduzzi
TitleBuilding an Open-Source Cross-Cloud DevOps stack for a CRM Enterprise Application: A Case Study
AfilliationCommunication Systems
Project(s)MELODIC: Multi-cloud Execution-ware for Large-scale Optimised Data-Intensive Computing
StatusPublished
Publication TypeProceedings, refereed
Year of Publication2019
Conference NameIFIP International Conference on Open Source Systems
Pagination3-11
Date Published05/2019
PublisherSpringer
Abstract

Open Source software solutions play a critical role for the SMEs by enabling easy access to reusable software. Also, with the rapid growth in the popularity of the cloud technologies, computational demands of SMEs are cost-efficiently met by the public clouds as users can dynamically acquire resources on demand according to their needs. However, non-standardized cloud interfaces, lack of inter-cloud transparency, and complex cost models, often result in vendor lock-in. Once in vendor lock-in, cloud users have to live with a single cloud provider and accept whatever pricing schemes and SLAs are imposed. Moreover, new regulations covered by the General Data Protection Regulation (GDPR) in Europe require companies to enforce policies regarding secure storage of data in the cloud, as well as restrict moving confidential datasets outside Europe. This situation requires a more transparent use of cloud resources from multiple cloud providers, that conform with user’s data privacy needs, service requirements, and budget.

In this paper, we discuss and pitfalls of designing a Cross-Cloud stack for an app-based extension platform of a Customer Relationship Management (CRM) system. The fully-automated DevOps stack, based on open source software tools and technologies, has been developed in close coordination with an open source integration project, Melodic. With the help of our DevOps , third-party apps in our CRM software are now Multi-Cloud ready, and the data storage in the cloud by the users conforms to potential GDPR requirements. In addition, the deployment time of apps has been reduced to minutes, while the platform is able to scale up and scale down apps efficiently based on the current workload requirements, saving substantial cloud costs.

Citation Key26465