UPOZORNĚNÍ: Jsou dostupné studijní plány pro následující akademický rok.

Multicore CPU Computing

The course is not on the list Without time-table
Code Completion Credits Range Language
NIE-MCC Z,ZK 5 2P+1C English
Garant předmětu:
Ivan Šimeček, Tomáš Vondra
Daniel Langr, Tomáš Vondra
Daniel Langr, Tomáš Vondra
Department of Computer Systems

Students will get acquainted in detail with hardware support and programming technologies for the creation of parallel multithreaded computations on multicore processors with shared and virtually shared memory, which are today the most common computing nodes of powerful computer systems. Students will gain knowledge of architecturally specific optimization techniques used to reduce the decrease in computing power due to the widening performance gap between the computational requirements of multi-core CPUs and memory interface throughput. On specific non-trivial multithreaded programs, students will also learn the basics of the art of creating these applications.


Basics of programming in C and C ++ (at the level of subjects BIE-PA1 and BIE-PA2), completion of the subject Parallel and Distributed Programming (MI-PDP). Recommended completion of the course Effective Programming in C ++ (MIE-EPC; formerly MIE-MPC).

Syllabus of lectures:

1. Multiprocessor, multicore and multithreaded architectures.

2.-3. (2) Sequence code optimization.

4. Models of relaxed consistency of shared memory.

5. Optimization of synchronization operations in multithreaded algorithms.

6. Memory allocators for parallel calculations.

7.-8. (2) Lock-free and wait-free programming and data structures.

9.-10. (2) Thread planning and working with memory in NUMA architectures.

11.-12. (2) Case studies of multithreaded algorithms.

13. Multithreaded programming and vectorization.

Syllabus of tutorials:

In computer exercises, students solve problems related to lectures; the instructor provides consultations and advice.

Study Objective:
Study materials:

1. Solihin, Y. : Fundamentals of Parallel Multicore Architectures (1st Edition). Chapman & Hall/CRC Computational Science, 2015. ISBN 9781482211184.

2. Sorin, D. J. - Hill, M. D. - Wood, D. A. : A Primer on Memory Consistency and Cache Coherence. Morgan & Claypool Publishers, 2012. ISBN 1608455645.

3. Pllana, S. - Xhafa, F. (Eds) : Programming Multicore and Many-core Computing Systems. Wiley, 2017. ISBN 0470936908.

Further information:
No time-table has been prepared for this course
The course is a part of the following study plans:
Data valid to 2023-06-01
Aktualizace výše uvedených informací naleznete na adrese https://bilakniha.cvut.cz/en/predmet6628806.html