Question 1

What file formats are supported?

Accepted Answer

CSV, JSON arrays, JSON lines and Parquet. Compressed files are handled directly, covering gzip, bzip2 and zstd, so there is no need to decompress before loading. That range covers most of what lands in a bucket, whether it is an export from another system, a partner feed or the output of a job. Parquet in particular carries its own schema, which makes typing more reliable than delimited formats.

Question 2

How do files map to tables?

Accepted Answer

Several ways, depending on how the bucket is laid out. A single file can become one table. Many files matching a pattern can be combined into one table, which is the usual arrangement for partitioned exports. Tables can also be discovered from the structure. Glob patterns are the useful part: values captured from the key path, such as a date or region folder, become columns on the resulting table, so partitioning that exists only in the path survives into the data warehouse.

Question 3

Can loading be triggered when files arrive?

Accepted Answer

Yes. Alongside scheduled runs, you can configure bucket event notifications delivered through a queue, so ingestion is driven by files landing rather than by a clock. That suits pipelines where files arrive irregularly, since it avoids both waiting for the next scheduled run and polling frequently for nothing. Scheduled runs remain available and the two approaches can be combined depending on the dataset.

Question 4

Why route files through a pipeline rather than loading them directly?

Accepted Answer

Snowflake can read from object storage itself, so the case here is about the surrounding work. A managed ETL pipeline handles schema creation and migration, turns path values into columns without writing extraction logic, applies format and compression handling uniformly, and keeps the load in the same place as the rest of your syncs with the same scheduling, monitoring and alerting. For a single well defined feed the difference is small; across many feeds maintained by different teams it adds up.

Question 5

How do I connect Amazon S3 to Snowflake?

Accepted Answer

Connect Amazon S3 with an access key and secret or an assumed role, granting read access to the bucket and prefixes you need. Connect Snowflake with account details, a role, a warehouse and credentials, with no separate storage bucket required since loads go through Snowflake's internal stage. Define the file patterns that make up each table, confirm the captured path columns look right, then set a schedule or configure event notifications to start the ETL sync.

Sync Amazon S3 to Snowflake

Sync Amazon S3 to Snowflake with Polytomic

CSV, JSON and Parquet

Path values become columns

Event driven option

Or sync Snowflake to Amazon S3

Trusted by businesses worldwide

Step-by-step integration guide

Connect Amazon S3 in Polytomic

Connect Snowflake in Polytomic

Create a sync between Amazon S3 and Snowflake

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 Amazon S3 with Snowflake?