Unity Catalog Managed Tables 2026: PM Insights at DAIS

0:04Hey Elizabeth, thanks for being with us. So, tell me more. Like, what are you working on? >> I'm working on managed tables and really excited about all the new things that are happening this year with managed tables. >> Yeah, I heard a man like tons of things are coming on managed tables. So, maybe can we start by defining what's a managed table and maybe how it's working, when can I use them? >> Absolutely. So, managed tables are

0:26different than other table types in the sense that Databricks maintains them automatically for you. Whereas with external tables, you have to take care of those automatic optimizations all by yourself. >> So, the external tables, that would be the one they leave outside in one bucket. I define the bucket and then I manage everything. >> Yeah, so we've got there's foreign tables, there's Unity Catalog external

0:48tables, and then there's Unity Catalog managed tables. Managed tables are the only ones where Unity Catalog actually has full visibility into all the recent writes and that because Unity Catalog can reliably see all those recent writes, that's what allows us to provide those automatic optimizations really safely, correctly, and accurately to lay out your data in the right way, to clean up data in the

1:11background, ultimately having helping customers to have 20 times faster queries and 50% lower cost, reducing kind of all this manual effort. >> Sounds amazing. Do you want to talk more about the all the optimization we can do like all the options we can have on the managed table like why why should we be using them? >> Sure. We need to make sure your life is easier. So, for example, we've got automatic liquid clustering that

1:34automatically lays out the data for you. Whereas with Unity Catalog external tables, you might have heard of something called liquid clustering. We have to select all those particular clustering keys, but that still takes a lot of work. And so, we what we do with managed tables is we automatically select those keys for you. And that changes as your data changes. >> on the queries, you see see who's using them and how they are using them and then we change

1:56the layout for you. >> Exactly. Exactly. So then your queries are a lot faster and we also compress files in the background as well. Um so that helps you save on storage costs. So, customers really like that aspect as well. >> And what's the what's the latest? >> What's the latest? We've got so many new features. We've We've got automatic upgrades so you get all the latest and greatest features on managed table with automatically. We have automatic time to

2:18live so you have you get if you want to set uh you know a column uh timestamp column, you can automatically delete rows of data after a certain period of time. Um and also managed iceberg tables are now generally available which is super exciting cuz that means now on iceberg tables as well as Delta, you get vector search, you can sync to lakehouse, you get data profiling, um

2:40and automatic change data feed. So now even at this particular summit uh and in general, we've got all these new uh capabilities on managed iceberg tables which is great. >> Do you have a session on managed tables? >> Yeah, I have a session on on uh Unity Catalog tables today. Um it's just later today. So I'm >> Oh, later today. Well, so folks have to join that one. We'll put the link in the comments.

3:02>> Absolutely. >> All right. >> Yeah. >> Well, thank you so much. And guys, you just go and put everything on managed table. >> Yeah. >> In lakehouse, >> drive data seeds, lake flow streams the flow. Iceberg and Delta lake, tables aglow. Unity Catalog governs lake base. OLTP strong. Genie takes to SQL. MOSAIC KO PROLONGS. YOU sipping quick in cohost

3:24[music] like >> the fire. Genie dreams scale higher and higher. Next gen drop it. [music]