Delavnica: Delo s superračunalnikom HPCFS (Working with supercomputer HPCFS)
ZOOM (udeleženci dobijo povezavo ob registraciji)
Opis: Na delavnici se bodo udeleženci spoznali z uporabo superračunalnika HPCFS. Predstavljeno bo delovanje superračunalnika, dostop do HPCFS, pregledali bomo seznam razpoložljive programske opreme. Priporočeno je, da udeleženci delavnice predhodno pridobijo uporabniški račun za HPCFS.
Superračunalnik, ki je na voljo na UL FS (imenujemo ga HPC FS) je sestavljen iz dveh računskih sklopov. Sklop z imenom haswell, kupljen v 2016, vsebuje 20 računskih vozlišč z INTEL Xeon E5-2680 v3 procesorji, kar je skupno 480 računskih jeder. Sklop rome, kupljen v 2021, vsebuje 20 računskih vozlišč s procesorji AMD Epyc 7402, kar je skupno 1920 računskih jeder. Pomembne komponente HPCFS so še vozlišča z grafičnimi pospeševalniki (2x), prijavna vozlišča (2x) in datotečni strežniki LUSTRE in NFS s skupnim neto spominom 271 TB. Posamezne komponente so med sabo povezane z IB QDR povezavo
Zahtevnost: Osnovna
Jezik: Glede na prijave
Termin: 27. 03. 2025 od 14.00 - 15.30
Virtualna lokacija: ZOOM
Ciljna publika: Zaposleni na Fakulteti za strojništvo, raziskovalci, študenti in projektni sodelavci, ki imajo pred začetkom delavnice odobren dostop do HPC.
Priporočeno predhodno znanje: Osnove dela v Linux ukazni vrstici (Bash, SSH)
Potek delavnice: Virtualno - na tematiko dela s superračunalnikom. Uporabljen bo tudi oddaljen dostop do HPCFS.
Na izobraževanju pridobljena znanja:
- Povezovanje s superračunalnikom preko NoMachine in SSH
- Prenos in shranjevanje podatkov na superračunalniku
- Kako zagnati in uporabljati programsko opremo na superračunalniku
Description: The workshop will introduce participants to the use of the HPCFS supercomputer. The operation of the supercomputer, access to HPCFS, and a list of available software will be presented. It is recommended that workshop participants obtain a user account for HPCFS in advance.
The supercomputer available at UL FS (we call it HPC FS) consists of two computing clusters. The cluster named haswell, purchased in 2016, contains 20 computing nodes with INTEL Xeon E5-2680 v3 processors, which is a total of 480 computing cores. The rome cluster, purchased in 2021, contains 20 computing nodes with AMD Epyc 7402 processors, which is a total of 1920 computing cores. Important components of HPCFS are also nodes with graphics accelerators (2x), login nodes (2x), and LUSTRE and NFS file servers with a total net memory of 271 TB. Individual components are interconnected by IB QDR connection
Difficulty: Basic
Language: Depending on applications
Date: 27. 03. 2025 from 14.00 - 15.30
Virtual location: ZOOM
Target audience: Employees at the Faculty of Mechanical Engineering, researchers, students and project collaborators who have approved access to HPC before the start of the workshop.
Recommended prior knowledge: Basics of working in the Linux command line (Bash, SSH)
Workshop course: Virtual - on the topic of working with a supercomputer. Remote access to HPCFS will also be used.
Knowledge acquired during the training:
Connecting to a supercomputer via NoMachine and SSH
Transferring and storing data on a supercomputer
How to run and use software on a supercomputer
Organizator:
Predavatelji:
Dr. Pavel Tomšič
He is a research assistant at ULFME and is well qualified for several HPC related topics. He is actively involved in efforts to raise competencies in the field of supercomputing, such as the Partnership for Advanced Computing in Europe (PRACE). He is also coordinator of Erasmus + project SCtrain - a strategic partnership for the transfer of knowledge from supercomputing between Slovenia, Austria, the Czech Republic and Italy. As part of the EuroHPC project for the establishment of European National Competence Centers in the field of supercomputing (EuroCC), he is the champion for Training and Skills Developmentfor NCC Slovenia.
E-naslov: pavel.tomsic@fs.uni-lj.si
Dr. Leon Kos
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 keydeveloper of PRACE MOOC Managing Big Data with R and Hadoop. He has been the leaderof PRACE Summer of HPC trainings in 2014, 2015, 2016, 2017, 2018 and 2019. He is also Slovenian holder of several national and international projects.
E-naslov: leon.kos@fs.uni-lj
Grega Tomažin
E-naslov: grega.tomazin@fs.uni-lj.si
