Advanced and Parallel Python

Performance

You’ve probably heard by now that Python is an easy to use, multi-purpose language but that it’s slow. While this might be true implementing algorithms naively, its performance can often be very good, and even on par with a C implementation.

The goal of this workshop is to give you the ability to take existing Python code and make it go faster. Way faster.

Topics

  1. The Scientific Python Software Stack
  2. Why (and What) is Python?
  3. Compiling Python Code
  4. Finding Bottlenecks
  5. Parallel Programming Concepts
  6. Using Multiple Cores
  7. Scaling Beyond One Machine

Credits

Workshop developed by Laurent Duchesne, originally based on McGill’s Advanced Parallel Python workshop.

Reference web site built using tools provided by Software Carpentry.