Scalable and Extensible Cloud-Based Low-Code Model Repository

Low-code development platforms (LCDPs) are becoming increasingly common in the soft-
ware industry. By leveraging visual diagrams, dynamic graphical user interfaces, and
declarative languages, these platforms support the development of full-fledged applications
in the cloud. However, given the rapid evolution of these platforms, they encounter a fleet of
challenges and limitations. To address the challenges in Low-Code Development Platforms,
it’s essential to study their core concepts and technologies, primarily Model-Driven Engineer-
ing (MDE) and cloud computing. Despite MDE’s progress, its broader adoption is hindered
by challenges faced by practitioners. The first obstacle is efficient support for discovering
and reusing existing model artifacts. The development of similar tools and extensions leads
to resource wastage, undermining productivity and collaboration in model-based processes.
Additionally, local deployment of modeling environments causes scalability, extensibility,
collaboration, and performance challenges. Consequently, modelers are required to engage
in a process that involves downloading both artifacts and executables to their local ma-
chines. This is a prerequisite step before initiating a potentially intricate and lengthy setup of
Model-Driven Engineering (MDE) tools prior to their effective utilization.

Throughout this dissertation, we attempted to advance state-of-the-art toward understanding
and supporting cloud-based modeling in terms of LCDPs. Therefore, we aimed to enhance
the scalability and extensibility of modeling infrastructures by developing a cloud-based
low-code model repository. This approach goes beyond the typical implementation of
repositories with simple storage and query capabilities. We provide a large-scale repository
and services for low-code engineering (LCE). The implemented repository enables access,
persistence, discovery, and reuse of modeling artifacts via scalable and extensible approaches
and infrastructures. In the LCE context, core services are containerized, orchestrated, and
deployed as cloud services. The repository’s functionalities can be extended via its remote
API or by adding functionality in the form of extensions and services. Finally, an integrated
web-based search platform and various domain-specific languages are devised to support
various mechanisms for composing, discovering, and reusing persisted artifacts and model
management services.