7 Comments

I really found it useful 😁

Expand full comment
Nov 18Liked by Zach Wilson

Great post. The dos and donts is my favorite part.

Expand full comment

This was super cool to learn about, even as someone mostly focused in frontend land. Thanks, Zach

Expand full comment

Seems to me at 100TB Spark will take a lot of time reading the data source before in-memory processing begins. How is this startup time accounted for?

Expand full comment

Very appreciative of all the DE’s continuously finding ways to access the data for analysis easier.

Those were such hairy tables to deal with. I had to deal with so many different reachability analysis that hit these tables joining them to other equally hairy advertiser tables.

Expand full comment

Thanks for the great post. It's not every time we would get to work at that scale, so it's nice to get the distilled findings.

Expand full comment