This program is designed to model simple implicit parallelism using Python's concurrent.futures.ThreadPoolExecutor. The program demonstrates how multi-threading can be used to speed up the process of summing a large array. It achieves this by dividing the work into chunks and processing them concurrently.
- Utilizes multi-threading to achieve parallelism.
- Divides a large array into chunks and processes them in parallel.
- The program generates an array of integers from 1 to 1,000,000.
- This array is divided into a thread count quantity of equal-sized chunks.
- Using
ThreadPoolExecutor, each chunk is processed concurrently by a separate thread to calculate its sum. - The chunk sums are then combined to form the total sum.
Run the .sln file in Visual Studio (2022), or
Install the dependencies and run the script in testingParallelism.py using Python cmd:
python thread_test.py
The program will start, and you’ll see detailed logs of the execution steps.
To run without logs, change log level from INFO to WARN.
Project derived from learnings from my reading of Game Engine Architecture 3rd Ed. By Jason Gregory, Chapter 4.