Infopaket

Fundamentals of distributed computing systems

ВБ3.1

Вибіркова дисципліна для ОП

Second

2021/2022

2 Semester

6

1 Know the basic principles of distributed computing systems 2. Know the basic principles of computer clusters based on the Linux operating system 3. Know the basics of the MPI protocol. 4. To know the basic methods of data transmission in the MPI protocol 5. Know the basics of OpenMP 6. Know the main approaches to program parallelization 7. To be able to organize calculations on clusters with different task management systems 8. Be able to write basic programs with data exchange between processes in MPI 9. Be able to write programs using collective data exchange and data reduction 10 To be able to parallelize programs using OpenMP 12. To be able to use parallel algorithms in problems of mathematical physics

Distance form

1. Have the basics of programming in the C (or C++) language. 2. To be able to apply previous knowledge of numerical methods. 3. Basic knowledge of higher mathematics courses is required (linear algebra, mathematical analysis, differential equations, methods of mathematical physics). 4. Necessary basic knowledge of theoretical physics courses (classical mechanics, electrodynamics, quantum mechanics, statistical physics)

The regulatory discipline "Fundamentals of distributed computing systems" is a component of the cycle of professional training of specialists of the educational and qualification level "Master of Physics". The course program is aimed at students who are already familiar with basic mathematical and physical disciplines. Knowledge of programming and numerical modeling methods is mandatory

1. Ortega J.M., Voigt R.G., Solution of partial differential equations on vector and parallel computers SIAM, 1985 2. Bertsekas D.P., Tsitsiklis J.N. Parallel and distributed computation.. numerical methods, Athena Scientific, 1997 3. Rauber T., Runger G. Parallel programming.. for multicore and cluster systems, Springer, 2010 4. Shonkwiler R., Lefton L. An introduction to parallel and vector scientific computing, CUP, 2006

The total volume is 180 hours1, including: Lectures – 30 hours. Practical classes - 30 hours. Independent work - 120 hours

- semester assessment 2nd semester: 1. Modular control work RN 1.3, 2.1 (10 points). 2. Modular control work of RN (10 points). 3. Tasks, independent work (40 points). A student is not allowed to take credit if he scored less than 36 points during the semester. final assessment in the form of credit. You can get a maximum of 40 points on the test. Conditions for admission to credit: solution of at least 30% of the tasks assigned for independent work.

Ukrainian