The main goal is to present to the students »computer friendly« writing of programs. Mainly we are focusing on better use of computer resources for faster computer performance. Contents: Parallelism on the instruction level ILP, VLIW, EPI, Evaluation of programming loops (for, do, repeat), data dependencies in loops, legal transformations, Methods of evaluation of data dependencies in loops (approximate, exact), Hierarchical test for data dependencies of loops, Programming of superscalar architecture computer, instruction scheduling, evaluations of loops, Local level programming, loop reorganisation, loop optimisation, Multilevel cache impact, Branching impact, Programming of multithreading architectures, SIMD instruction set (programming of SIMD architectures, programming language MMC), Operations in floating-point, EXAMPLES: optimisation of Intel (Pentium) processors, High- performance programming of typical problems of linear algebra, and signal processing of on-line frame signals in assembly, and language C, Introduction of Intel's programming libraries, Examples in languages Fortran and Intel C compilers, Programming tools for optimisations, debugging, IntelVtune, Typical Benchmarks for measuring computer performance.
** The Course is not available in the academic year 2012/2013.