HPC - Algorithms and Applications - Winter 13
- Term
- Winter 13/14
- Lecturer
- Prof. Dr. Michael Bader
- Time and Place
- Lecture: Monday, 14.00-15.30, MI 02.07.023 (starts Oct 21);
Tutorial: Wednesday, 10-12, MI 02.07.023 (starts Oct 23, roughly bi-weekly) - Audience
- Elective topic in Informatics Bachelor/Master: students in mathematics or in any science or engineering discipline are welcome!
- Tutorials
- Oliver Meister
- Exam
- written exam, Wednesday, Feb 5, 2014; 10-12 in room MI 02.07.023 (time and room of the tutorial)
a repeat exam (oral) will be offered on April 2 and 3 - Semesterwochenstunden / ECTS Credits
- 3 SWS (2V + 1Ü) / 4 ECTS
- TUMonline
- https://campus.tum.de/tumonline/lv.detail?clvnr=950111465 (lecture)
https://campus.tum.de/tumonline/wbStpModHB.detailPage?&pKnotenNr=705979 (module description)
Contents
Annonuncements
- From Nov 18, the lecture on Monday will start at 14.00 (instead of 14.15)
Content
The lecture will have a focus on parallel algorithms and implementation techniques in the field of numerical simulation and high performance computing, such as:
- linear algebra problems on dense and sparse matrices
- simulation on structured and unstructured meshes
- particle-based simulations (with long-range and short-range interactions)
- spectral methods (parallel FFT and related algorithms)
- Monte Carlo and statistical methods
(a.k.a. the seven dwarfs of HPC).
The accompanying tutorials will include practical assignments, and will concentrate on the programming of GPU and accelerator platforms.
Lecture Material
Slides and exercise sheets/solutions will be made available during the lecture.
Lecture slides will be published here after the lessons: See also the lecture from winter term 2012/13.
- Oct 21: Intro
- Oct 21, Oct 28, Nov 4: Fundamentals - Parallel Architectures, Models, and Languages
- read the related paper: Roofline: An Insightful Visual Performance Model for Floating-Point Programs and Multicore Architectures (technical report by Williams et al.)
- MPI examples: Cannon's Algorithm mpi_cannon.c (unsafe send/receive), mpi_cannon_sr.c (using MPI_Sendrecv), mpi_cannon_nbl.c (non-blocking communication)
- Oct 28, Nov 4, Nov 11: Dwarf No. 1 - Dense Linear Algebra;
- read the related paper: article by Elmroth et al. in SIAM Review
- chapter In-core dense matrices of the ScaLAPACK User's Guide
- Nov 18: Dwarf no. 2 - Sparse Linear Algebra: Application example (page rank) and data structures
- Nov 25, Dec 2: Parallel Sparse Matrix-Vector Multiplication
- lecture material accompanying the book by R. Bisseling (see esp. the slides psc4_3.pdf, psc4_4.pdf, psc4_5.pdf, and psc4_6.pdf)
- Dec 9, Dec 16: Dwarf No. 5 - Structured Grids
- articles by M. Frigo and V. Strumpen:
Cache oblivious stencil operations (preprint);
The memory behavior of cache oblivious stencil operations (preprint can be found via Google) - article by K. Datta et al. in SIAM Review (preprint)
- articles by M. Frigo and V. Strumpen:
- Dec 16, Dec 18: Structured Grids and Space-filling Curves (will not be part of the exam)
- Jan 13: Dwarf No. 6 - Unstructured Grids and Partitioning
- Jan 20, Jan 27: Dwarf no. 4: N-body methods and implementation
- Maple worksheet: twobody.mw (also as PDF)
- article on Fast Multipole methods by C.R. Anderson
- article by Barnes & Hut in Nature (both articles can be accessed via TUM ebib-access)
- Feb 3: "all questions answered" (exam preparation)
Tutorials
Roughly every second week a two hour tutorial will take place (details at page top; days and time will be announced in TUMonline and in the lectures). The assignments and their solutions will be gradually posted here.
Date | Slides | Worksheet | Source | Source (solution) |
Oct 23rd | Organizational remarks | - | - | - |
Nov 6th | Introduction to CUDA | Worksheet 1 | Exercise 1 | Solution 1 |
Nov 13th | Further details on Dense LA in CUDA | Worksheet 2 | Exercise 2 | Solution 2 |
Nov 27th | Sparse LA in CUDA | Worksheet 3 | Exercise 3 | Solution 3 |
Dec 11th | Solving the heat equation with CUDA | Worksheet 4 | Exercise 4 | Solution 4 |
Jan 8th | The Shallow Water Equations and CUDA | Worksheet 5 | Exercise 5 | Solution 5 |
Jan 22nd | Further topics on SWE and CUDA | Worksheet 6 | - | - |
Exam
- Repeat Exam: will be scheduled as oral exam on April 2/3; the same rules as for the written exam will apply (esp. no helping material, questions on the tutorials)
- no helping material of any kind will be allowed for the exam
- please make sure that you register for the exam in TUMonline
- the exam will extend over all topics discussed in the lectures and tutorials:
- approx. 30% of the questions will deal with questions related to the tutorials; basic knowledge about GPU programming with CUDA is thus necessary
- The following topics will be excluded as topics of the exam:
Literature and Online Material
- R.H. Bisseling: Parallel Scientific Computing - A structured approach using BSP and MPI, Oxford University Press, 2004.
- Course notes on Rob Bisseling's lecture on Parallel Algorithms (based on the text book)
- V. Eijkhout: Introduction to High-Performance Scientific Computing (textbook, available as PDF on the website)
- T.G. Mattson, B.A. Sanders, B.L. Massingill: Patterns for Parallel Programming, Addison-Wesley, 2005
- G. Hager, G. Wellein: Introduction to High Performance Computing for Scientists and Engineers, Chapman & Hall/CRC Computational Science, 2010
Books on CUDA
- D.B. Kirk, W.W. Hwu: Programming Massively Parallel Processors - A Hands-on Approach, Morgan-Kaufman, 2010
- J. Sanders, E. Kandrot: CUDA by Example, Addison-Wesley, 2011
Prerequisites
Helpful, but not strictly required is knowledge in:
- basics of numerical methods (e.g.: lecture IN0019 Numerical Programming or similar)
- basics of parallel programming (lecture Parallel Programming, HPC - Programming Paradigms and Scalability, or similar)
Most important is a certain interest in problems from scientific computing and numerical simulation!