Python for HPC

This course teaches strategies to write well-performing Python code suitable for HPC applications.

Event details

When

Nov 20, 2018 10:00 AM to
Nov 21, 2018 05:00 PM

Where

MPCDF, IPP/D2

Contact Name

Add event to calendar

Python for HPC, November 2018, MPCDF, Garching

 

Abstract


In recent years the Python programming language has become more and more popular in scientific computing for various reasons. Users not only implement prototypes for numerical experiments on small scales, but also develop parallel production codes, thereby partly replacing compiled languages such as C, C++, and Fortran. However, when following this approach it is crucial to pay special attention to performance. This course teaches approaches to use Python efficiently and reasonably in a HPC environment. The first lecture gives a whirlwind tour through the Python programming language and the standard library. In the following, the lectures strongly focus on performance-related topics such as NumPy, Cython, compiled C- and Fortran extensions, profiling of Python and compiled code, parallelism using multiprocessing and mpi4py, and efficient IO with HDF5. In addition, we will cover topics more related to software-engineering such as packaging, publishing, testing, and the semi-automated generation of documentation. Finally, basic visualization tasks using matplotlib and similar packages are discussed. The lectures will be given based on Jupyter notebooks and will include many reusable code examples. For each topic and after each lecture, a hands-on exercise is planned, therefore participants are encouraged to bring a laptop. On the second day, there will be the opportunity for the participants to apply the strategies taught to their codes with support from MPCDF staff.

Lecturers


  • Sebastian Ohlmann, MPCDF
  • Klaus Reuter, MPCDF

Venue


The course will take place on November 20 and 21 at the IPP Garching, building D2, in the lecture hall and/or in the seminar room in the second floor.

  • 2018/11/20, 10:00 to 17:00: Lectures and exercises.
  • 2018/11/21: 10:00 to 17:00: Lectures and exercises, hands-on and code support.

Target Audience


  • Computational scientists who run Python code on MPCDF HPC systems
  • Scientists interested in Python programming with a focus on performance

Registration


Update 2018/10/16: As a large number of participants has registered very quickly, it is now only possible to register for the first day (Nov 20) when the course will take place in the big lecture hall in building D2. On the second day we have the restriction to all fit into a seminar room.
Please register with your email address here. Only participants affiliated with a Max Planck institute are eligible.

Document Actions