Topic outline

• General
  

  General

     

  This course, a collaboration of Erlangen Regional Computing Center (RRZE) and Leibniz Supercomputing Center (LRZ), is targeted at students and scientists with interest in programming modern HPC hardware, specifically the large scale parallel computing systems available in Jülich, Stuttgart and Munich, but also smaller clusters in Tier-2 centers and departments.
  
  The exercises given below are suggestions - you are free to play with any code you like. If you are a beginner you will probably also not be able to complete all tasks.
  
  Note:

  • Topics 1-5 below refer to the March 6 - 10 course phase (taking place at RRZE)
  • Topics 6-9 refer to the April 3 - 6 course phase ("Advanced Topics", taking place at LRZ)

  Course room:

  RRZE Building, University Campus Erlangen, Martensstr. 3, 91058 Erlangen, room 02.135
  

  Contact:

  Dr. Georg Hager (georg.hager@fau.de)
  Dr. Volker Weinberg (weinberg@lrz.de)

  • Hotel information for Erlangen Page
  • Schedule for days 1-5 File
  • Useful links and references Page
• Day 1
  

  Day 1

  The lectures on day 1 cover general tools and practices that are useful for working in programming environments, and specifically when dealing with HPC systems.

  • Welcome and general remarks File
  • Introduction to HPC File
  • Programming Environment @ LRZ File
  • Using the CoolMUC-3 Cluster at LRZ File
  • Secure connections with SSH File
  • Exercise: Working with the LRZ SuperMIC system Page
  • Exercise: Working with the LRZ CoolMUC-3 system Page
  • Environment Modules File
  • User Interfaces (screen & vnc) File
  • Introduction to GNU make File
  • Exercise: Using GNU make Page
  • Make Tar File with Sources
  • Managing software development with SVN and git File
• Day 2
  

  Day 2

  Day 2 gives an introduction to the dominant HPC programming models (MPI and OpenMP) as well as to computer architecture on a general level.

  • Elementary MPI and OpenMP File
  • Exercise: π by the Monte Carlo method Page
  • Introduction to computer architecture File
  • Affinity by OpenMP places File
  • Exercise: Measuring the FP divide throughput Page
• Day 3
  

  Day 3

  On day 3 we cover the principles of serial code optimization, and in the afternoon we get an in-depth look at OpenMP.

  • Principles of serial code optimization File
  • Exercise: General code optimization Page
  • Exercise: OpenMP-parallel vector triad Page
  • Advanced MPI Programming File
  • Exercise: Parallelization of a raytracer code Page
  • Spoiler alert: Solution for MPI File
• Day 4
  

  Day 4

  The lectures of day 4 also introduce parallelism on a theoretical level (parallel patterns, performance models) and the different variants and structures of parallel computer systems.

  MPI as the dominant distributed-memory parallel programming model is also covered in depth on day 4.

  • Parallel architectures in depth File
  • Parallel Algorithms File
  • Advanced OpenMP Programming File
  • Exercise: Parallel histogram computation Page
  • Exercise: Eliminating recursion Page
  • OpenMP performance and correctness File
  • Evening event at RRZE Page
  • Talk: Thirteen modern ways to fool the masses with performance results on parallel computers File
• Day 5
  

  Day 5

  Day 5 is deals with Intel tools for performance analysis, tuning and correctness checking of parallel programs. It also introduces the Scalasca tuning framework.

  • Basics of Software Engineering File
  • Scalasca File
  • Score-P File
  • Performance analysis tools: Intel® Application Performance Snapshot and ITAC File
  • Performance analysis tools: Intel® VTuneTM Amplifier and Advisor File
  • Timetable advanced course File
  • HPC Training at LRZ File
  • Some solutions to exercises File
  • Exercises as archive File
• Day 6
  

  Day 6

  PRACE Course: Advanced Topics in High Performance Computing

  This workshop is a PRACE Training Course courtesy the agreement of the Gauss Supercomputing Centre with PRACE with respect to Europe-wide HPC training. LRZ is obliged to PRACE for partial support of the costs for this event.

  

  Day A1 of the advanced topics course deals with node-level code optimisation.

  • Infos Advanced Course 2018 File
  • Schedule for days A1-A4 (advanced topics) 2018 File
  • Using the CoolMUC-3 Cluster at LRZ File
  • Info: Working with the LRZ CoolMUC-3 system Page
  • Architecture Specific Optimization Part 1 File
  • Exercise 1: The stream benchmark Page
  • Architecture Specific Optimisations Part 2 File
  • Architecture Specific Optimisations Part 3 File
• Day 7
  

  Day 7

  Day A2 deals with MPI-I/O, I/O on LRZ's HPC systems, as well as introducing specialised I/O libraries.

  • Parallel I/O with MPI IO File
  • Exercises for MPI IO Page
  • Tuning I/O on LRZ's HPC systems File
  • I/O Profiling with the Darshan Tool File
  • Exercises I/O Profiling with Darshan File
• Day 8
  

  Day 8

  Day A3 continues with I/O libraries and provides lectures about Intel Xeon Phi programming.

  • SIONLib Library File
  • High-Level I/O Library: HDF5 File
  • I/O libraries Exercises File
  • Introduction into Intel KNL Programming File
  • Exercise on KNL memory and cluster modes File
• Day 9
  

  Day 9

  Parallel programming with PGAS languages
  

  • PGAS programming with Coarray Fortran and Unified Parallel C File
  • Exercises File

    including how to compile and run PGAS programs

  • Closing Remarks File
• Evaluation
  

  Evaluation

  Please fill out the course evaluation form before leaving the lecture room to help us and PRACE to increase the quality of PRACE trainings.

  https://events.prace-ri.eu/event/675/evaluation/evaluate

  https://goo.gl/aayoAn
  

  Thank you for your collaboration and thank you for joining our course!!