Scaling Data Science with Modin

on YouTube ↗46m

" summary

Presentation explains how to scale Data Science workflows using Modin, a drop-in accelerated DataFrame library compatible with pandas. It describes using Modin to speed up and parallelize pandas-based data processing for larger datasets and higher performance without changing existing pandas code.

AI-written summary · about

" chapters (16)

1. 0:00Welcome
2. 1:48Why Modin today
3. 3:35Pandas standard
4. 6:31Pandas big data issues
5. 8:36Prototype to production gap
6. 10:23What is Modin
7. 17:15Modin for data scientists
8. 23:01Modin technical details
9. 27:22Modin parallelism engines
10. 31:57Pandas replacement
11. 33:02Modin cluster scalability
12. 34:27Cloud cluster tasks
13. 36:13Dask and SQL server integration
14. 39:53Modin performance with Dask SQL
15. 40:59Modin key benefits
16. 41:43Q&A

" related talks

• 21mScaling Interactive Data Science with Modin and RayDevin Petersohn · Ray Summit 2021
• 24mData Scientists are More Valuable than Hardware: The Design of ModinDevin Petersohn · Ray Summit 2020
• 29mLearning to Scale Data Science, Machine Learning, and Pandas with Ray and ModinDevin Petersohn · PyData Los Angeles 2018
• 39mAccelerated Data Science LibrariesAshot Vardanian · PyData Yerevan 2022
• 01h 24mData Science at Scale with DaskRichard Pelgrim · PyData London 2022
• 36mScaling data workloads using the best of both worlds: pandas and SparkEng & Kwon · PyData Seattle 2023
• 01h 23mData Science at Scale with DaskRichard Pelgrim · PyCon DE & PyData Berlin 2022
• 34mRAPIDS: How to accelerate your data science pipeline by orders of magnitudeAdam Grzywaczewski · PyData Eindhoven 2020