Open Source Platform for Scalable Multi-Purpose Virtual Desktop Infrastructures

Data and access to it are central to each NFDI consortium. However, moving data around is often impractical—it may be too large, sensitive, or restricted by agreements with, e.g., the funding provider, and copying introduces duplication, versioning issues, and loss of provenance. Rather than bringing data to the researcher, a Desktop-as-a-Service (DaaS) approach can offer researchers interactive, high-performance access in a secure and efficient manner. Driven by the need for seamless workflows and efficient data handling in NFDI4BIOIMAGE, we present a DaaS approach that is broadly applicable across NFDI. It supports diverse use cases, such as standardized virtual training environments for distributed participants (like required in DataPLANT), remote visualization of large-scale HPC datasets, and secure access to sensitive data (BERD)—all without the overhead of local machine setup and maintenance.