The DeiC Front Office at DTU supports the national HPC infrastructures.

The DeiC Front Office

DTU has annual access to HPC resources through DeiC national HPC Infrastructures and from 2022 to resources provide via EuroHPC and LUMI.

****************************************************************
Please sign up to the HPC channel at DTU
https://teams.microsoft.com/l/team/19:EGeJnjgrTfCH_w30wBo7VrDh1oj_1aIQmJJbzuNvEi01@thread.tacv2/conversations?groupId=1362d5bc-3593-4319-bea1-4990471c15a0&tenantId=f251f123-c9ce-448e-9277-34bb285911d9

... or enter the following code to get access to the latest news around HPC at DTU in Teams
6l6qgb1
****************************************************************


DTU has annual access to HPC resources through DeiC national HPC Infrastructures and from 2022 to resources provided via EuroHPC and LUMI. DeiC HPC is divided into various levels of computational resources.

Interactive HPC (formerly known as Type I) focuses on interactive computing resources and easy access for new users.

Typically, the individual researcher's computer is insufficient due to computing power, storage, or memory. HPC Systems with many cores and the ability to handle substantial amounts of data is mentioned as Throughput HPC (formerly known as Type II).

Three Throughput HPC systems are available at a national level for DeiC applicants
• Computerome 2.0 (Xeon Gold 6230 Cascade Lake, 40 cores, 192 GB). Read more ... https://www.computerome.dk/display/C2W/Computerome+2.0+Wiki
• GenomeDK (AMD EPYC Rome 7452, 64 cores, 512 GB). Read more ... https://genome.au.dk/
• Sophia (AMD EPYC 7351, 32 cores, 128 GB). Read more ... https://dtu-sophia.github.io/docs/

Large Memory HPC (formerly known as Type III) is characterized by large memory (currently four TB) with a few cores—usage on computational cases that cannot be distributed between computer notes. Accelerated HPC and Capability HPC (formerly described as Type IV and V) is not yet available as described on The DeiC HPC systems' site. The national HPC resources provided via DeiC are described on https://www.deic.dk/en/supercomputing.

We are a team with a broad understanding of the national High-Performance Computing (HPC) infrastructures provided by DeiC. We will assist in helping to get the right resource for your research. Furthermore, we can assist in choosing the best solution for you through a high knowledge of HPC.
Applying for resources on any HPC system submitting a grant application for a research project is necessary.

LUMI provides access to GPU and AMD based GPU resources.

Contact deic-frontoffice@dtu.dk