Introducing Bodo Community Edition (for Pandas acceleration)

Hi everyone,

Some of you may have heard about bodo.ai (https://bodo.ai), a python supercomputing analytics platform built with Numba as its compiler infrastructure. Bodo enables automatic parallelization and optimization of data analytics code with the simple addition of a @bodo.jit decorator, and provides up to 100x faster performance than Spark.

I’m excited to introduce the Bodo Community Edition, which allows free use of Bodo on up to four cores. It can accelerate Pandas code up to 100x in our testing (UDFs, strings, date/time typically benefit the most). Try it out with a simple Conda install (2. Installation — Bodo 2021.2.2+8.g16038f75 documentation), and let me know what you think! Your feedback will help us make Bodo better for the community.

Here are some additional resources:
YouTube : bodoai - YouTube
Blog: bodo.ai

if you want to have Bodo specific conversations, there’s a discourse you can join: https://discourse.bodo.ai/
as well as a community Slack : https://join.slack.com/t/bodocommunity/shared_invite/zt-lhvq3s8h-EFgce0gKGK1T1BNVO3KdxQ
and a GitHub for more detailed feedback, requesting features and reporting bugs: GitHub - Bodo-inc/Feedback

5 Likes

Hello @ehsantn could you tabular side by side comparison with Intel SDC as I am not sure what features are offered by both, especially the Types supported etc

@scheung38 you can look at the documentations to compare API coverage:
https://docs.bodo.ai/latest/source/pandas.html
https://intelpython.github.io/sdc-doc/latest/apireference.html