GitHub user melin edited a comment on the discussion: Support Internal Catalog

In China, data development personnel generally do not pay much attention to 
whether to use a data lake. They are more focused on real-time data being 
written to the data warehouse. A data lake is needed. There is no such high 
degree of substitution in the foreign market as using iceberg to replace hive. 
Only some ODS tables use Paimon or iceberg. The main ones still use Parquet 
format tables. Using a data lake will also increase maintenance costs and 
storage costs. It is relatively difficult to implement using iceberg or Paimon 
for all tables. 

Technically, an HMS is required. If the customer requests a cloud-native 
solution, they may question why an HMS is still needed (believing that only 
Hadoop would require an HMS). It is hoped to find an alternative solution to 
HMS. It is believed that more people have such demands.

GitHub link: 
https://github.com/apache/gravitino/discussions/7652#discussioncomment-13729823

----
This is an automatically sent email for [email protected].
To unsubscribe, please send an email to: [email protected]

Reply via email to