DKF - HPC documentation
Welcome to the Komondor HPC System
The Komondor supercomputer, operated and continuously developed by Digital Government Development and Project Management Ltd. (DKF), provides high-performance computing (HPC) resources to support scientific research, technological development and innovation. The service is available to a broad range of users, including universities and research institutions, large enterprises, small and medium-sized enterprises (SMEs), and public sector organisations.
Designed to accommodate diverse computational workloads, Komondor offers a range of computing environments, including CPU and GPU clusters, a large-memory node with 12 TB of RAM, and an AI partition featuring 8-GPU nodes for multi-GPU computing and AI workloads. These resources enable users to tackle computationally demanding challenges, from scientific simulations and engineering applications to AI development and other advanced computing tasks.
Komondor is accessible to users across Hungary through the high-bandwidth, low-latency HBONE+ research network, with harmonised user authentication facilitating access. The system also provides a multi-tier storage infrastructure, including high-capacity storage options to support computational workloads and the data they generate.
Komondor is Hungary’s largest national high-performance computing system, providing a powerful platform for tackling computationally demanding challenges across academia, industry and the public sector. Users can take advantage of a wide range of development tools and scientific applications, as well as visualisation tools for analysing and presenting computational results.
Access to the service is available to users affiliated with eligible organisations that have an established legal relationship with DKF, subject to registration and the applicable access conditions.
If your organisation is interested in using Komondor, contact us at hpc@dkf.hu to discuss your computing needs and explore the available options.
Note
The documentation is regularly updated to reflect the latest information. If you have any questions or feedback, please contact us at hpc-support@dkf.hu.
Table of contents
- HPC Echo
- Calls and Deadlines
- First steps
- System Overview
- Storage
- Basic Usage
- Overview
- Preparing Job Scripts (sbatch)
- Submitting and Managing Jobs
- Checking Job Efficiency
- Troubleshooting Resource Efficiency
- Interactive Use
- Exposing Web Services with
frp - Resource Usage
- Quality of Service (QOS)
- Advanced Job Definitions
- Data Sharing for Project Members
- File Management Considerations
- Software
- Overview of Installed Software
- Environment Modules
- Container Environment
- Installing Python packages
- Conda environments with Apptainer
- EESSI
- EasyBuild (EB)
- Bio Software using EESSI
- Alphafold2
- Genomic Analyses
- Ansys Fluent, CFX
- OpenFOAM
- Chem Software using EESSI
- ABINIT
- AMS/ADF
- AMBER
- CFOUR
- CHARMM
- CP2K
- DL_POLY
- GROMACS
- Molpro
- MRCC
- NAMD
- OPENMolcas
- ORCA
- Q-Chem & BrianQC
- TeraChem
- Geosciences using EESSI
- Mathematics / Numerical Methods / Optimization using EESSI
- MATLAB
- Wolfram Mathematica
- Jupyter Environment
- Cray Programming Environment
- AI/deep learning
- Programming Models
- HPC Support
- How to cite
- FAQ
- Courses
- EU Supporting Projects