The training is designed as an introductory course of High Performance Computing (HPC), reflecting the supercomputing infrastructure at the IT4Innovations National Supercomputing Center (IT4I).
Its syllabus is divided into six parts:
- HPC systems architecture, IT4I infrastructure intro
- Accessing and working on the IT4I clusters
- Code development on the cluster – an overview for application users
- Parallel programming basics
- Efficient system usage – profiling, tuning, optimization
- Development of HPC architectures – path to exascale
Please note that the course timetable is preliminary and is subject to change.
Target audience and Purpose of the course
This training is dedicated to the employees and students at the Institute of Molecular and Translational Medicine of the Palacký University Olomouc, who are interested in basics of HPC and (potential) approach to the IT4Innovations supercomputing infrastructure. The audience focused on are primarily application users with some experience from smaller platforms (desktops), who need to take advantage of HPC to solve their problems in larger scale, which is out of scope of their current resources. Typically scientists from various application fields, who are not developers of the application employed, but need to understand better what is “behind the scene”, e.g. how parallelization works.
The course will provide them with the introduction to most important aspects of HPC, especially as practiced at IT4I, and help answer related questions.
About the tutor(s)
The presenters are researchers at the IT4Innovations National Supercomputing Center and/or Ph.D. students at VSB – Technical University of Ostrava:
- Martin Beseda received his M.Sc. in Computational Mathematics from VSB – Technical University of Ostrava. After few years of industrial experience he got a position at IT4I in the Parallel Algorithms Research Lab as a developer of scientific software. Currently he is focused on research in the field of Quantum Chemistry and finishes his Ph.D. study in Computational Science.
- Jakub Beránek is a researcher and a computer science Ph.D. student who is working as a research assistant in the Advanced Data Analysis and Simulations Lab. He is interested in distributed systems, code optimization and profiling, GPUs and applied machine learning.
- Ondřej Meca is a researcher in the Infrastructure Research Lab. His research interests include verification of parallel algorithms and development of scalable pre/post-processing algorithms for large-scale engineering problems.
- Radim Vavřík is a Ph.D. student in Computational Science and a research assistant in the Infrastructure Research Lab. He is mainly interested in parallel computing, scalable algorithms design, GPU acceleration and code optimization, and heterogeneous architectures. He worked on hydrological and flood modeling software and high-performance heterogeneous platform for energy-efficient computing. Now, he works on GPU acceleration of the ESPRESO library.
- Ondřej Vysocký received his M.Sc. degree in Computer Science from the Brno University of Technology, Czech Republic in 2016. His master thesis focused on parallel I/O optimization. Currently he is a Ph.D. student at VSB – Technical University of Ostrava and at the same time he works at IT4I in the Infrastructure Research Lab. His research is focused on energy-efficiency in high performance computing. He was an investigator of the Horizon 2020 READEX project which dealt with energy efficiency of HPC applications using dynamic tuning. Since that time, he develops a MERIC library, tool for energy measurement and hardware parameters tuning during a parallel application run. He is also member of the PowerStack initiative, which works on a holistic, extensible and scalable approach of power management.
This course/event was supported by The Ministry of Education, Youth and Sports from the Large Infrastructures for Research, Experimental Development and Innovations project „e-INFRA CZ– LM2018140“.