Skip to main content
Skip header

HPC libraries and tools

Type of study Doctoral
Language of instruction Czech
Code 9600-0003/01
Abbreviation HPCKN
Course title HPC libraries and tools
Credits 10
Coordinating department IT4Innovations
Course coordinator prof. Ing. Tomáš Kozubek, Ph.D.

Subject syllabus

The course consists of deeper introduction to currently used libraries and tools for the development of parallel applications. The libraries are mainly selected from fields such as linear algebra, accelerated computations including the tools accelerating the development of applications. With respect to the computational technologies, the selected libraries will cover programming for the distributed memory systems, systems with shared memory and multi-core processors, and systems with accelerators.

- Libraries for linear algebra: BLAS, LAPACK, ScaLAPACK, Plasma
- Libraries for accelerators: MAGMA, Nvidia CUDA libraries
- Tools for accelerating the development of applications: PETSc, Intel MKL

Literature

1. Gene H. Golub and Charles F. Van Loan, Matrix Computations (Johns Hopkins Studies in the Mathematical Sciences), Dec 27, 2012

Advised literature

1. Scientific articles describing computational methods used in selected libraries