cu-dbmi / demo-python-memory-reference-accounting Goto Github PK

Explore built-in Python capabilities through added tests covering (for example):

• Reference counting challenges
• Large(er) data / memory scenarios
• Places where garbage collection may be triggered by a larger amount of time

This issue highlights the possibility of exploring how memory consumption works alongside threaded or parallel process management within Python. Ideas here:

• Explore built-in Python multithreading and multiprocess memory usage
• Explore non-built-in Python multithreading and multiprocess memory usage
• Explore this also through multiple memory allocators (i.e. non-built-in capabilities)

Explore non-built-in Python aspects of memory through added tests covering (for example):

• NumPy memory management
• Pandas memory management
• PyArrow memory management
• How multiple memory allocators share (or don't share) memory resources
• Contextualize how the Python garbage collector works in the above