วีดีโอ

how do you manage schema changes? that was not demoed here.

Really good explanation of Iceberg.

Thank you so much for this presentation, is there any way i can get the slides?

Thank you. Great presentation and captured real world scenarios well

Great episode, awesome speaker!

Thanks Dan! This is one of the best talks I have listened to on Iceberg implementation. Automated table maintenance is the real deal.

Hi, may I ask like who stores these puffin files, or rather where are they stored. I am basically trying to Connect Spark with Iceberg, I am a bit confused about how to figure out or find the puffin files if I want to. Can you please provide an explanation if possible?

Great presentation. How are you orchestrating maintenance tasks? Is this on a naive schedule or event based?

Great Presentation! You mentioned that there is some sort of compaction, cleanup etc. but what I not yet get who is doing those housekeeping tasks? Is it the catalog who performs maintenance or is this something the ingesting parties do?

Super candid to call out the “undifferentiated work”

You can use Apache Nessie, it a modern catalogue with versioning capabilities.

Great Explanation!

awesome talk!

Amazing explanation!!!

There should be a huge asterisk next to the aforementioned REST catalog. It's not free or open source. The only good production ready catalog out there is nessie. Which Daniel doesn't mention (I guess because dremio are tabular's competitors).

Does this mean we dont need an additional transform job to do the upsert/merge into once the kafka sink pushes the data into iceberg table? Is the merge into handled by kafka sink and populates the final target table with no additional code?

is REST Catalog project still in progress. I could not find any official releases or documentations about how to deploy it on premise. thanks

I failed to create nested y/m/d partition for iceberg table in Athena, how to accomplish this?

I have a question around S3 lifecycle cleanup. Let's say I want to keep only a month worth of data. I could put a lifecycle policy on the data files for a month, but the issue is I now have orphaned data files in the manifest lists. Is the only way to call the expensive delete orphan operation?

Nice job Thomas! Way to go! 👍

Way to go son

First viewer, really it is a good data lakehouse platform

Hi. Thanks for the demo video. I'm keen to try out the library for typical read | write | remove | upserting data (incl. table management as you already demonstrated). From a documentation perspective the project seems fresh, so please excuse if I'm running ahead with my question... Does the library support any writing functionality to tables at the moment? (could not see it from documentation, or after installing the pyiceberg lib locally and looking at the functions exposed after loading a table)

How does tabular stand out from Hive, AWS Glue Catalog and others?