Nuvolos basic concepts

Nuvolos brings computation and project management together in one place. Researchers run analyses, educators deliver courses, and teams collaborate — all within a platform designed so that data stays under institutional control, workloads scale on demand, environments remain stable across sessions, and results are reproducible end to end.

A handful of core ideas make this possible. Understanding how they fit together will help you get the most out of Nuvolos.

Everything lives in a tree

All work in Nuvolos is organised into a hierarchy: organisations contain spaces, spaces contain instances, and instances hold the actual working material — files, data, and applications. An organisation typically maps to a real-world entity (a university, a department), a space maps to a project or course, and instances are parallel branches of work within that project. This tree structure determines who can see what, who pays for what, and how content flows between people.

Work can be frozen and restored

Any instance can be snapshotted at any time. A snapshot captures the complete state — files, tables, application configurations — as an immutable, time-stamped record. You can restore from a snapshot to roll back, or use it as a baseline for a new experiment. Snapshots are the mechanism behind versioning, reproducibility, and safe experimentation on the platform.

Content can be moved around

Nuvolos provides a distribution mechanism for pushing content — individual files, tables, applications, or entire snapshots — from one location to another. Distribution powers many workflows: delivering course material to students, onboarding a colleague into a research project, or sharing a reproducible environment with referees. Applications can also be exported as Docker images for use outside of Nuvolos.

Access is controlled by roles

A role system governs what each user can do at every level of the tree. Organisation managers oversee all spaces; space administrators control membership and can distribute content; Instance Editors modify the working state; Instance Viewers have read-only access. Spaces can further be set to public, faculty-only, or private visibility. This layered model makes it possible to run anything from an open dataset to a confidential research project within the same organisation.

Resources are tracked through resource pools

Compute time, storage, and credits are tracked through resource pools that map onto the tree at the organisation or space level. This lets institutions allocate separate budgets for different projects, departments, or grants. For details, see Resources and the hierarchy.