pg8000 VS quokka

As others have said you can use a custom compiled version of the lib, lambda layer or use lambda images, however, if you're not committed to psycogp2 I've found pg8000 a much easier library to work with in Lambda. You can just install it as any other library without any problems.

When people say "pure X", to me, it normally means they didn't involve an FFI or external compiler. This is an often beneficial thing since it simplifies your build process.

For example, here [0] is a "pure Python postgres driver" and the implication is that it doesn't use libpg.

Or see also this discussion [1].

[0] https://github.com/tlocke/pg8000

[1] https://www.reddit.com/r/learnpython/comments/nktut1/eli5_th...

I agree that information shouldn't be duplicated, but in one of my projects I've taken the opposite approach and made the FAQ the only place that certain information is presented. It's for the library https://github.com/tlocke/pg8000 and I've called them 'Examples' rather than a FAQ, but each time I get a question that isn't covered by the examples I add in a new example. I'd be interested to hear what people think of this approach.

An awesome read!

Something related that I found out about from HN a few months back is another engine called quokka. It's particularly interesting and applicable how quokka schedules distributed queries to outperform Spark https://github.com/marsupialtail/quokka/blob/master/blog/why...

Hi Justin, you might be interested in my blog: https://github.com/marsupialtail/quokka/blob/master/blog/bac... advocating a cloud based approach.

You don't have to use the system I am building, but it's worth thinking about that design.

SQL support is very challenging.

I work on Quokka (https://github.com/marsupialtail/quokka). I support Iceberg reads. Recently we are adding SQL support from just parsing the DuckDB logical plan, though that is very challenging as well.

The Python world lacks a standard for a plug and play SQL query optimizer. Apache Calcite is good for the JVM world, but not great if you are trying to cut out the JVM.

In case people are interested, I wrote a post about fault tolerance strategies of data systems like Spark and Flink: https://github.com/marsupialtail/quokka/blob/master/blog/fau...

The key difference here is that these systems don't store data, so fault tolerance means recovering within a query instead of not losing data.

would love to collaborate on an integration with pyquokka (https://github.com/marsupialtail/quokka) once I put out a stable release end of this month :-)

Then you should keep an eye on quokka. This may become the "Spark" for Polars/DuckDB. It seems to be under active development though I'm not sure how stable it is.