quinn VS spark-rapids

We're brainstorming functions to make PySpark easier, see this issue: https://github.com/MrPowers/quinn/issues/83

Adding some README documentation to the README should be quite straightforward. Here's a function that needs to be documented: https://github.com/MrPowers/quinn/issues/52 .

quinn is a library with PySpark helper functions. I need to work through all the open issues / PRs and bump all versions. I should do another release. This library gets around 600,000 monthly downloads.

Pandas syntax is far inferior to regular PySpark in my opinion. Goes to show how much data analysts value a syntax that they're already familiar with. Pandas syntax makes it harder to reason about queries, abstract DataFrame transformations, etc. I've authored some popular PySpark libraries like quinn and chispa and am not excited to add Pandas syntax support, haha.

Here's how I added a create_df method to the SparkSession class: https://github.com/MrPowers/quinn/blob/main/quinn/extensions/spark_session_ext.py

They spent a lot of effort improving the catalyst engine under the hood too and making it easier to extend and improve it in the future. Making it easy to add your own native code to Spark itself. Shameless plug of a blog post I wrote on this subject which basically reiterates what Matthew Powers, author of Spark Daria and quinn, wrote here.

I built daria (https://github.com/MrPowers/spark-daria) to make it easier to write Spark and spark-fast-tests (https://github.com/MrPowers/spark-fast-tests) to provide a good testing workflow.

quinn (https://github.com/MrPowers/quinn) and chispa (https://github.com/MrPowers/chispa) are the PySpark equivalents.

Built bebe (https://github.com/MrPowers/bebe) to expose the Spark Catalyst expressions that aren't exposed to the Scala / Python APIs.

Also build spark-sbt.g8 to create a Spark project with a single command: https://github.com/MrPowers/spark-sbt.g8

I've built popular PySpark (quinn, chispa) and Scala Spark (spark-daria, spark-fast-tests) libraries.

His newer project, Ballista, was also donated to Apache Arrow. I hope to get the Rust skills to collaborate with him on open source work someday too. He's also doing really cool work on spark-rapids FYI.

The spark-rapids project allows one to run multi-GPU ETL workloads on a Spark cluster. https://github.com/NVIDIA/spark-rapids In such a setup, the GPU nodes are part of the Spark cluster. Multi-GPU nodes are viable, although an executor is currently limited to a single GPU.

So, in my day job at NVIDIA, I work on the RAPIDS Accelerator for Apache Spark, which is an open-source plugin that provides GPU-acceleration for ETL workloads, leveraging the RAPIDS cuDF GPU DataFrame library.