Metoder inom högprestandaberäkningar
Logga in till din kurswebb
Du är inte inloggad på KTH så innehållet är inte anpassat efter dina val.
Kurswebben har varit stängt för redigering sedan första juni 2025 och kommer stängas ned helt första oktober 2026. Utifrån hur kurswebben har använts finns det några olika alternativ för ersättare:
- Sidan "Inför kursval" i Om kursen
- Kurs-PM (i Om kursen)
- Publikt utrymme i Canvas
Är du intresserad av mer information om detta, kontakta e-learning@kth.se.
DD2356 - Methods in High-Performance Computing
What is this Course About?
- Making effective use of the most powerful parallel supercomputers for problems in science and engineering
- Doing this requires paying attention to every part of the parallel system
- It also requires a scientific and rigorous approach to performance
Course Goals
- Understand the sources of performance in modern architectures
- How to establish performance expectations
- How to diagnose performance problems
- How to design algorithms and applications to maximize performance potential
- How to exploit parallelism at all levels
Intended Learning Outcomes
At the end of this course, you will be able to:
- Describe and list HPC architecture design choices and features
- Connect to the supercomputers, build and run codes on multicore systems and clusters
- Use HPC tools to measure and analyze the performance of computer codes running on supercomputers
- Design, implement and optimize a parallel engineering application for HPC systems using OpenMP and MPI
Prerequisites
- Knowledge of Linux (Supercomputers have Linux OS)
- Knowledge of C/C++ or Fortran - Assignments are in C/C++ though
- Knowledge of command line editors (emacs, vi, ...)
Course Structure
- Pre-recorded video lectures
- In-person lectures
- Assignments
- Quizzes
- Final project
