Selinix opened a new issue, #11618:
URL: https://github.com/apache/iceberg/issues/11618
### Query engine
Spark for loading, Trino for querying
### Question
Hi,
I’m looking for guidance on the most efficient solution for maintaining full
history and querying the latest versions of events without maintaining
redundant copies of the data.
A use case is to be able to query either:
1. All versions of an event (e.g., `SELECT * FROM full_hist WHERE id =
'XXX'`)
2. Only the latest version of an event (e.g., `SELECT * FROM latest_slice
WHERE id = 'XXX'`)
The latest version is determined by the maximum value in a `version` field
for each `id`.
**Questions**
1. Is it better to maintain:
- A single table with full history and periodically deduplicate it into a
separate `latest_slice` table?
- Or a single full history table with a view that computes the latest
versions dynamically?
2. If the latter, does applying optimization techniques like partitioning,
sorting, and ordering on the full history table significantly improve
performance for querying the latest versions?
3. Given the preference to store only one copy of the data, what is the most
performant and practical solution for this scenario?
Thank you for your guidance!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For additional commands, e-mail: issues-h...@iceberg.apache.org
