Question 1

What write modes are available?

Accepted Answer

Create, update, update or create, and replace. The first three need an identity field so rows can be matched. Replace is the one to understand before selecting it: it replaces all rows in the destination rather than deleting the specific rows missing from your query, so it suits a table that Polytomic owns outright and not one that other systems also write to. Update or create is classified into insert and update work rather than issuing a single database upsert.

Question 2

Is data staged or written directly?

Accepted Answer

It depends on the sync type, and the difference matters. A bulk sync replicating tables stages rows into a scratch table and then merges into the final table, so readers never see a partially written result. A model sync writes directly with no staging table, batching inserts up to 250 rows at a time and issuing updates one statement per row. For a large modelled table refreshed frequently, that distinction is worth planning around.

Question 3

Does the MySQL table need a primary key?

Accepted Answer

Not for a model sync. You map an identity field from the Snowflake model onto a destination column and updates match on it, whether or not MySQL has a key declared on that column. Uniqueness still matters, since a non unique identity will update every matching row. For bulk table replication the picture is different: with a primary key rows merge on that key, and without one the documented behaviour is recreating or truncating and reloading the table.

Question 4

What drives cost on the Snowflake side?

Accepted Answer

Compute on the warehouse running the query, since reads are ordinary SQL rather than an unload. Selecting fewer columns, filtering in the model, and maintaining a good tracking column all reduce how much gets scanned. A dedicated warehouse with auto suspend is the usual recommendation so sync activity does not compete with analyst queries and does not sit idle billing you. Model syncs are differential after the first backfill, which keeps ongoing runs small.

Connect Snowflake with MySQL

Connect Snowflake and MySQL with Polytomic

Schema handled automatically

Four write behaviours

Staged loads on bulk syncs

Trusted by businesses worldwide

Step-by-step integration guide

Connect Snowflake in Polytomic

Connect MySQL in Polytomic

Create a sync between Snowflake 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 Snowflake with MySQL?