Python for Finance: Analyze Big Financial Data

Figure 8-4. Execution speed depending on the number of threads used (eight-core machine)

EASY PARALLELIZATION

Many problems in finance allow for the application of simple parallelization techniques, for example, when no

data is shared between instances of an algorithm. The multiprocessing module of Python allows us to efficiently

harness the power of modern hardware architectures without in general changing the basic algorithms and/or

Python functions to be parallelized.