Delavnica: Parallel programming with MPI / OpenMP

To exploit large massively parallel cluster paradigms combining MPI and OpenMP is used. Moreover MPI and OpenMP standards are evolving including new ideas and features to become increasingly effective in new machines. This gives developers of HPC applications a smooth path of evolution of their applications without having to deal with heavy re-factoring to take up new technologies.

The 2-day course will cover topics related to parallelism, OpenMP tasks, the OpenMP memory model, performance tuning, hybrid OpenMP + MPI and OpenMP implementations. The course is aimed at programmers seeking to deepen their understanding of OpenMP.

The course is delivered in an intensive two-day format using UL-FME’s training facilities. It is taught using a variety of methods including formal lectures, practical exercises, programming examples and informal tutorial discussions. After the course the participants should be able to write more efficient OpenMP programs.

Udeleženci/Attendees
SLO: Število prijav je omejeno na 30.

ENG: The target audience consists of postgraduate students and young researchers of natural and technical sciences, engineers from industry where supercomputing can be used as competitive advantage (automotive, electronic, material industry), logistics, etc.
The number of applicants is limited to 30.

Predhodnje znanje/ Prerequisite knowledge
ENG: For the hands-on sessions participants should know how to work on the Unix/Linux command line and have intermediate skills in programming with C/C++. Since the focus of the school is on parallelization, participants have to be familiar with the topic and must have basic knowledge of OpenMP and MPI

Pridobljena znanja / Skills to be gained
SLO:
V pripravi

ENG:
Understand the message passing model
Implement standard message passing algorithms in MPI
Debug simple MPI codes
Measure and comment on the performance of MPI codes
Understanding of best practice for MPI+OpenMP programming.

Predavatelji/About the authors
Janez Povh, PhD
He is an associate professor at ULFME with a long track of teaching duties at university (a former Dean at Faculty of Information studies in Novo Mesto, Slovenia). With mathematical background he is a specialist in Big Data methods and their implementation in Hadoop and RHadoop. Beside this he is a well experienced with creating and implementing parallel algorithms for mathematical optimization problems. He is one of the leading educators in PRACE MOOC Managing Big Data with R and Hadoop. He is also Slovenian holder of several national and international projects.

Leon Kos, PhD
He is an assistant professor at ULFME and is well qualified for several HPC related topics. He is a qualified trainer from the HLRS train-the-trainers program and was the key developer of PRACE MOOC Managing Big Data with R and Hadoop. He has been the leader of PRACE Summer of HPC trainings in 2014, 2015, 2016, 2017, 2018 and 2019. He is also Slovenian holder of several national and international projects.

Leon Bogdanovic, MSc
He is an MSc student and software developer at LECAD lab ULFME and a former technical collaborator at ULFMST. He has experience in developing accident reconstruction/simulation and GPU accelerated plasma physics simulation codes. His current interest are highly parallel codes for plasma field-line tracing in tokamak type fusion reactors.

Ivona Vasileska, MSc
She is an assistant and researcher at ULFME and also PhD student in field of nuclear engineering in ULFMF. Since 2016 she has been worked on kinetic and fluid plasma modelling using different kind of codes for it. Mostly she has experienced in Particle In Cell (PIC) kinetic codes and SOLPS-ITER fluid code. Also in 2020 she was a mentor in PRACE Summer of HPC school. The topic of that project was make a transfer form CPU to GPU of the PIC code. She has more than 12 publication and has been working on several project in fusion and HPC computing.

Borut Cerne, MSc
He is an assistant researcher, working at ULFME. His research work has so far mainly been focused on finite element method based thermomechanical modelling of homogeneous and composite polymers, used in gearing applications. Apart from analytical/numerical modelling he has substantial experience in experimental testing of polymer and other non-metal materials. Currently, he is involved in several research and knowledge dissemination projects related to the application of HPCs in the field of mechanical engineering. Part of his duties working at ULFME also involve pedagogical work as a teaching assistant on several CAD/CAE related faculty subjects. Additionally, he was involved in various R&D oriented industrial projects.

---

Projekt EuroCC je financiran s sredstvi Skupnega podjetja za visoko zmogljivo računalništvo (EuroHPC JU) v skladu s sporazumom o dodelitvi sredstev št. 951732. EuroHPC JU je prejelo finančno podporo iz EU programa Obzorje 2020 ter Nemčije, Bolgarije, Avstrije, Hrvaške, Cipra, Češke , Danske, Estonije, Finske, Grčije, Madžarske, Irske, Italije, Litve, Latvije, Poljske, Portugalske, Romunije, Slovenije, Španije, Švedske, Združenega kraljestva, Francije, Nizozemske, Belgije, Luksemburga, Slovaške, Norveške, Švice, Turčije, Republike Severne Makedonije, Islandije in Črne gore.