Norwegian Research Infrastructure Services¶
The Norwegian academic high-performance computing and storage infrastructure is maintained by NRIS (formerly known as the Metacenter), which is a joint collaboration between UiO, UiB, NTNU, UiT, and Sigma2.
We provide valuable resources for the research communities. Not only do we provide state-of-the art compute and storage facilities, backed by support and a guarantee that your data always stays in Norway. But possibly more important is easy access to a wide selection of competences that can assist, realize or take your project to the next level.
Compute, storage, pre/post-processing, visualization, machine learning¶
We offer compute resources (Betzy, Fram, and Saga, storage resources (NIRD - National Infrastructure for Research Data), as well as the NIRD Toolkit platform for pre- and post-processing analysis, data intensive processing, visualization, artificial intelligence, and machine learning.
Researchers also have access to LUMI through the LUMI consortium.
First time on a supercomputer?¶
Please read the GETTING STARTED section (left sidebar). In the sidebar overview you will also find technical details about the machines, instructions for using installed software, for submitting jobs, storage, and code development.
Please do not hesitate to write to email@example.com if you find documentation sections which are not clear enough or have suggestions for improvements. Such a feedback is very important to us and will count.
How to get the most out of your allocation¶
We want to support researchers in getting the most out of the high-performance computing services. When supporting users, we see that these problems are very frequent:
Reusing outdated scripts from colleagues without adapting them to optimal parameters for the cluster at hand and thus leaving few cores idle. Please check at least how many cores there are on a particular cluster node.
Requesting too much memory which leads to longer queuing and less resource usage. Please check How to choose the right amount of memory.
Requesting more cores than the application can effectively use without studying the scaling of the application. You will get charged more than needed and others cannot run jobs. If others do this, your own jobs queue.
Submitting jobs to the wrong queue and then queuing longer than needed. Please take some time to study the different Job Types.
If you are unsure about these, please contact us via firstname.lastname@example.org and we will help you to use your allocated resources more efficiently so that you get your research results faster.
Acknowledging use of national HPC infrastructure¶
Projects are required to acknowledge the use of the national e-infrastructure resources in their scientific publications. Papers, presentations and other publications that feature work that relied on resources provided by Sigma2 should include an acknowledgement following this template.
Text is licensed CC-BY¶
Unless explicitly noted, all text on this website is made available under the Creative Commons Attribution license (CC-BY-4.0) with attribution to NRIS.