All researchers at AAU as well as students in relation to relevant projects when approved by supervisor.

No. In AI Cloud you are only allowed to work with public or internal information according to AAU’s data classification model (classified as levels 0 and 1, respectively).

A similar solution for working with confidentoinal or sensitive data (classified as levels 2 and 3) is being developed.

The AI Cloud front-end node is accessed from a terminal application found on your local computer that runs SSH (Secure Shell). Secure Shell is part of most modern operating systems, such as most distributions of Linux, macOS, and Windows 10 and above. Find out how to connect to AI Cloud on the AI Cloud documentation ‘Introduction’ page.

Yes - it is possible to share files among different users on the cluster.

In principle any software that can run under Linux since the operating system on the servers in AI Cloud is Linux. In many cases, it is necessary to encapsulate the software in a container in order to use it. The software must be able to utilise the (NVIDIA) GPUs in the platform. The GPUs cannot automatically accelerate software that was not designed for it.

No. As a rule of thumb, you can only operate applications through a command line interface. AI Cloud is best suited for throughput type computing: queue up your job and come back for the result later. If your project could benefit from running on a GPU cluster but demands an interactive graphical interface, contact CLAAUDIA research support at support@its.aau.dk.