National HPC facilities
The national HPC landscape consists of four different types of supercomputers and a user interface.
Type 1 - Interactive HPC
For type 1, the focus is on interactive computing resources and easy access for new users. In the national context, this type is a new type of facility that is targeted at users who don't have much experience with calculations on large facilities. Typically, when the individual researcher's own computer is not sufficient due to lack of computing power, storage or memory.
Experienced users can, among other things, use this type of system to work with R statistics, but also for prototyping and idea development, just as it could be the students' first encounter with HPC systems. It is expected that this type of HPC will help increase the use of HPC in research for a number of new users.
The facility is operated by a consortium consisting of SDU, AAU and AU. The SDU computer Abacus make up one part of the resources and the AAU computer CLAAUDIA make up the other part. It will be AU that handles the support and helps new users get started.
Type 1 is fully ready to receive users from March 2021, but is already receiving users to a lesser extent. Additional resources and tools will be expanded in the coming period.
Type 2 - Throughput HPC
This type of facility typically has a large number of cores which can be a mix between cost-effective and calculation-efficient units. Type 2 also has the ability to handle large amounts of data and focus on high security. In relation to calculations within health science, technical simulations, chemistry, physics and bioinformatics in a broad sense, there will often be requirements for HPC systems with a focus on high "throughput performance".
The Type 2 facility is operated by a consortium consisting of AU, DTU and KU. The computer resources are provided by Computerome 2, which is jointly owned by DTU and KU, by GenomeDK at Aarhus University, and by Sophia, which is run by DTU.
Type 2 is already ready to receive the first users.
Type 3 - Large Memory HPC
This type of HPC focuses on problem solving, with a structure that can't be easily or efficiently distributed between many computer nodes. This is a type of system that is characterized by typically relatively few cores with access to a large globally addressable memory area. The application is, for example, in certain areas of classical chemistry, physics, signal processing with, for example, handling of large matrix problems and quantum chemistry.
Type 3 is under construction and is expected to be able to receive the first users on 1 March 2021. The facility is being developed and will be operated by SDU.
Type 4 - Accelerated HPC & Capability HPC
Type 4 is a type of facility whose primary computing capacity comes from accelerators of various kinds. The primary purpose of establishing the Type 4 computer is to provide an opportunity to test alternative accelerator solutions and research the use of future HPC solutions. The system is not as such a production facility, but more an HPC laboratory. However, it will be open to researchers who can utilize this technology to perform computational tasks on the computer.
The Type 4 facility is under construction and is expected to be ready for operation at the end of 2021. The facility is being developed and will be operated by KU. Behind the project is a collaboration between ITU, RUC and KU.
Project 5 - development of a common access system
To support the construction of the national HPC landscape, DeiC's board has also launched a project, which right now goes by the name Project 5. The project's purpose is to create a unified online platform that will give researchers simple and secure access to the four computer types, as well as managing the resource distribution at the individual facilities.
Behind the project is a consortium with SDU, AAU and DTU, which will collaborate with DeiC and the parties behind the operation of the 4 types to develop the platform. The project starts with a workshop that will help determine the project.
The background of the national HPC landscape
The report Future national HPC landscape - recommendations from the Working Group for Future National HPC Landscape