Ruprecht-Karls-Universität Heidelberg
Siegel der Universität Heidelberg

Module for [Scientific Computing]

[back] to List of Modules.

[High-Performance Computing] - [2015 Sommer]

Module Code
IPHR
Name
High-Performance Computing
Credit Points
8 LP CP
Workload
240 h
Duration
ein Semester
Cycle
0
Methods Lecture 4 h + Exercise course 2 h
Objectives To have a firm command of relevant programming models for parallel high-performance computing and a knowledge of parallel algorithms in several application areas.
Content Introduction to parallel computer architectures Shared memory programming models, basic algorithms for shared memory: condition synchronization, locks, semaphores, cache coherence, multithreading, GPGPU programming Message-passing programming model: asynchronous vs. synchronous communication, message Passing Interface (MPI), basic algorithms for global communication: broadcast, reduce, barrier Remote procedure call and its object oriented extensions • Performance analysis of parallel algorithms • Applications: dense linear algebra, solution of sparse linear systems, particle methods, sorting
Learning outcomes Know the basic design of current parallel high-performance computers Understand the different programming models with their strengths and weaknesses Formulate and analyze parallel algorithms from several application areas
Prerequisitesnone
Suggested previous knowledge Programming in C++, IAD
Assessments weekly excercises, programming projects, written or oral exam
Literature
zum Seitenanfang