Accelerating object serialization by using constraints
How we achieved 3x-100x faster data serialization to a binary format or to JSON using low-level Cython and Python C API.
- Track: Python devroom
- Room: UD2.218A
- Day: Sunday
- Start: 09:30
- End: 10:00
- Video only: ud2218a
- Chat: Join the conversation!
The talk will demonstrate how doing less can result in doing more: how domain-specific data constraints may lead to dramatic speedups in serialization. There will be two examples: binary caching of numpy object arrays (Pandas) and generating json web responses from @dataclass-es. We will also cover the revolution of using C++17 in Cython, custom arena memory allocators based on mimalloc, and gory internals of CPython.
Speakers
Vadim Markovtsev |