SubjectsSubjects(version: 953)
Course, academic year 2023/2024
   Login via CAS
High Performance Computing in Physics - NFPL244
Title: High performance computing ve fyzice
Guaranteed by: Department of Condensed Matter Physics (32-KFKL)
Faculty: Faculty of Mathematics and Physics
Actual: from 2023
Semester: winter
E-Credits: 5
Hours per week, examination: winter s.:2/2, C+Ex [HT]
Capacity: unlimited
Min. number of students: unlimited
4EU+: no
Virtual mobility / capacity: no
State of the course: taught
Language: Czech
Teaching methods: full-time
Teaching methods: full-time
Guarantor: Ing. Dominik Legut, Ph.D.
doc. RNDr. Karel Carva, Ph.D.
Annotation -
High Performance Computing in Physics. General rules, basic methods how to work with sustem used at largest clusters of Czech national supercomputing center (IT4Innovations) and similar centres. Preparation of computationally intensive problems (optimization, parallelization), execution of jobs at computational clusters and other practical aspects. Recommended for bc. students of physics. Precondition is to be capable of basic handling of Unix/Linux systems.
Last update: Mikšová Kateřina, Mgr. (11.05.2023)
Course completion requirements -

To participate in the exam it is necessary to complete the credit. Credit requirements require active participation in exercises. For this reason corrective attempts to grant credit are not possible. The exam contains oral part, and its requirements follow the subject syllabus as presented during lectures.

Last update: Mikšová Kateřina, Mgr. (11.05.2023)
Literature -

1. Fortran 95/2003 Explained, Michael Metcalf, John Reid, and Malcolm Cohen, Oxford University Press (2004)

2. Numerical Recipes in Fortran 90 - The Art of Parallel Scientific Computing, William H. Press,

3. Saul A. Teukolsky, William T. Vetterling, Brian P. FlanneryPerformance Optimization of Numerically Intensive Codes, S. Goedecker, A. Hoisie, SIAM, Philadelphia, PA, 2001

4. J. Dongarra, J. Bunch, C. Moler and G. W. Stewart, Linpack User's Guide, SIAM, Philadelphia, PA, 1979, nebo online verze LAPACK Users' Guide

5. J. Dongarra, Sourcebook of parallel computing, Morgan Kaufmann Publishers, San Francisco, 2003

Last update: Mikšová Kateřina, Mgr. (11.05.2023)
Syllabus -

1. What is High Performance Computing. History plus how the development at IT4I goes..

X86 vs. GPU, vs. what is the compiler, parallelization etc.

2. Basic Unix commands, editing files by Nano, etc.

3. Scheduler, Modules, basic usages of batch system (Slurm, PBSPro) etc..

graphics visualization, detecting different nodes, etc.

4. Queues, detecting nodes, projects accounting, limits of queues, etc.

5. Easybuild and its control, how to use it and tailor it, making group and usage of module within the group

6. Libraries for mathematical operations (linear algebra packages) LAPACK, BLAS, etc.

7. Basics of Fortran

8. Makefile, compiling of the code (eg. Fortran), change and linking. Static vs. Dynamical libraries.

9. Basic of bash, loops for, while, basic math, etc.

10. Profiling of the code

11 -14 i) Practical parts - Quantum-Mechanical Calculations of Electronic structure of simple elements, equilibrium lattice constant, bulk modulus, demonstration of the power of prediction.

It’s possible to include problems proposed by students

Last update: Mikšová Kateřina, Mgr. (11.05.2023)
 
Charles University | Information system of Charles University | http://www.cuni.cz/UKEN-329.html