Hands-On Reactive Programming with Python

Using thread pools for CPU-bound operations

There are two possible ways to deal with CPU-bound operations:

• Split them into multiple processes
• Split them into multiple threads (provided that several execution engines are available on the CPU)

For now, the multithreading solution will be investigated. More details on how to solve it with multiprocessing will be provided in Chapter 10, Testing and Debugging.

As explained in Chapter 5, Concurrency and Parallelism in RxPY, two operators allow us to deal with schedulers and execution contexts: observe_on and subscribe_on. In the case of this application, ThreadPoolScheduler can be used. Since the encoding requests use a lot of CPU, and they can run in parallel, a thread pool can be used to execute ...