The challenge
There is no blueprint for adapting a REST (Representational State Transfer) server such as the EDC connector as a cloud-native serverless application. The software architecture of the EDC had to be analyzed in detail and a completely new “serverless architecture” developed. The goal was to reuse the basic EDC modules while taking advantage of commercial cloud technologies. However, the aim was not to replicate the power of the EDC project's extension mechanism, but rather to integrate adequate cloud services as standard according to the Pareto principle. New software engineering tools were created for the development to optimally support unit testing and continuous integration in a serverless project.
Our services
Compared to the original EDC, the “serverless EDC” was expanded to become a multi-tenant connector—a feature that is regularly missing in the original. This allows multiple participants in a data space or even an entire data space to be operated with just one deployment in the cloud. This expansion significantly reduces maintenance costs, especially when a large number of connectors are used, such as in an “intra-dataspace.”
The result
The “Serverless EDC” will make it easier to enter the world of data spaces in the future. No developer or programming skills are required for deployment within an AWS account, while the data provider retains full control over their installation – unlike with SaaS offerings. The same deployment of the Serverless EDC can therefore be used to implement both small demo installations and high-availability scenarios with huge data transfers. Thanks to the scalability of the serverless architecture and the available capacity of an AWS cloud, the serverless EDC can be used flexibly. The costs are mainly based on the actual use of the connector. Small data space scenarios can even be operated free of charge within the free AWS quota.