Question 1

What can I read out of Databricks?

Accepted Answer

Tables across your catalogs and schemas, or an arbitrary SQL query used as a model source. The second tends to be the more useful option here, since the shape a relational database wants is rarely the shape of a lakehouse table, and resolving that once in SQL is simpler than reshaping downstream. Where Unity Catalog is enabled, catalogs are discovered and referenced with three part naming, and system and internal schemas are hidden.

Question 2

What happens to nested values?

Accepted Answer

Databricks exposes STRUCT, ARRAY and MAP types, so the question is how you want them to land. STRUCT and ARRAY retain their nested schema where it can be parsed, while MAP is treated as a generic JSON shape. On a destination that cannot represent nesting, those values commonly arrive as strings, which is the most frequent first load surprise. Resolving them into scalar columns in the model SQL avoids that. DECIMAL keeps precision and scale where the destination allows.

Question 3

Does the MySQL table need a primary key?

Accepted Answer

Not for a model sync. You map an identity field from the Databricks model onto a destination column and updates match on it, whether or not MySQL has a key declared there. It should be unique, since a non unique identity will update every row it matches. Bulk table replication behaves differently: with a primary key it merges on that key, and without one the table is recreated or truncated and reloaded.

Question 4

Can reads be incremental?

Accepted Answer

Yes, using a tracking column, with the first run reading in full and later runs resuming past the stored value. A primary key is not required, though a unique key helps the reader chunk efficiently where many rows share the same tracking value. Tracking column reads do not detect deletes, and there is no change data feed reader on the source side, so a periodic full run is the way to reconcile if the MySQL copy needs to match exactly.

Question 5

How do I sync Databricks to MySQL?

Accepted Answer

Connect both systems first. Databricks needs a SQL warehouse hostname and a token, or an OAuth service principal, while MySQL needs host, port, database and credentials. Then define a model over a table or SQL query, map fields onto the MySQL table, and choose the write behaviour and identity field. Setting a tracking column at this point is what keeps later runs incremental rather than reading everything each time.

Connect Databricks with MySQL

Connect Databricks and MySQL with Polytomic

Tables or arbitrary SQL

Four write behaviours

Schema kept in step

Trusted by businesses worldwide

Step-by-step integration guide

Connect Databricks in Polytomic

Connect MySQL in Polytomic

Create a sync between Databricks and MySQL

Map fields and enable the sync

Frequently asked questions

The sync and ETL platform that does it all

Powerful data sync

Advanced analytics

Enterprise security

Easy setup

Ready to connect Databricks with MySQL?