Google BigQuery Destination

Released

All Stitch plans

Not applicable

Stitch supports the following Google Storage Location regions:

• United States (US)

• European Union (EU)

• London (europe-west2)

• Singapore (asia-southeast)

• Tokyo (asia-northeast)

• Sydney (australia-southeast)

Unsupported

Stitch does not support using SSH tunnels to connect to BigQuery destinations.

Supported

Stitch will attempt to use SSL to connect by default. No additional configuration is needed on your part.

Unsupported

Supported

To improve your query performance, you can apply partitioning and/or clustering to Stitch-created tables in your BigQuery destination. Your settings will remain intact even when new data is loaded.

Refer to the Table partitioning and clustering guide for application instructions.

• Connecting a Google BigQuery Destination

• Change the Location of a Google BigQuery Destination

Unsupported

BigQuery destinations do not currently support upserts, or updating existing rows.

When Stitch loads data into BigQuery destinations, it will load it in an append-only fashion. This means that new records and updates to existing records are appended to the end of tables as new rows. Existing records are never updated.

More info and examples can be found in our Replication Method documentation and Append-Only querying guide.

Unsupported

BigQuery destinations do not have native support for Primary Keys.

Unsupported

BigQuery destinations do not support multiple data types in a single column.

To accommodate this scenario, Stitch will create additional columns for each data type. Refer to the Table structural changes guide for examples and more info.

Supported

BigQuery destinations natively support nested data structures.

More info and examples can be found in BigQuery’s documentation.

No reported incompatibilities

No integration compatibility issues have been reported for BigQuery.

Insensitive

This means that BigQuery destinations treat upper- and lowercase characters as the same. For example: CUSTOMERID and customerid are equal in BigQuery.

When column names are loaded into Redshift, any uppercase characters will be forced into lowercase. This means that Id and id will canonicalize to the same name, which can lead to object collision errors. Refer to the Column naming section of this guide for examples.

1,024 characters

Tables with names that exceed this limit will be rejected. Rejected tables are logged in the _sdc_rejected table.

128 characters

Columns with names that exceed this limit will be rejected. Rejected columns are logged in the _sdc_rejected table.

10,000 columns

None

4 MB

Records that exceed this limit will be rejected. Rejected columns are logged in the _sdc_rejected table.

Unsupported

BigQuery will store the value in UTC as TIMESTAMP.

More info about timestamp data types can be found in BigQuery’s documentation.

0001-01-01 to 9999-12-31

Date values outside of this range will be rejected and logged in the sdc_rejected table.

More info about BigQuery date data types can be found in BigQuery’s documentation.

0001-01-01 00:00:00 to 9999-12-31 23:59:59.999999 UTC

Timestamp values outside of this range will be rejected and logged in the sdc_rejected table.

More info about BigQuery time data types can be found in BigQuery’s documentation.

-9,223,372,036,854,775,808 to 9,223,372,036,854,775,807

Integer values outside of this range will be rejected and logged in the _sdc_rejected table.

Up to 38 digits of precision

Decimal values outside of this range will be rejected and logged in the _sdc_rejected table.

Up to 9 digits of scale

Decimal values outside of this range will be rejected and logged in the _sdc_rejected table.

None

While there aren’t any size limits on string data, the record as a whole must still be less than 4 MB or it will be rejected and logged in the _sdc_rejected table.