Data types and features

This page provides an overview of the data types supported by different outputs. It also indicates whether or not the output supports upserts, aggregations, and SQL.

When creating an output, these specific data types are supported. Additionally, upserts, aggregations, and SQL are supported for most of the outputs.

Output	Data Types	Upserts	Aggregations	SQL
Athena	DOUBLE$^{6}$ BIGINT$^{6}$ BOOLEAN STRING TIMESTAMP DECIMAL(18,2) DECIMAL(18,5) DECIMAL(18,6) See: Note 6	+	+	+
Redshift Spectrum	DOUBLE$^{6}$ BIGINT$^{6}$ BOOLEAN BIGINT STRING TIMESTAMP DECIMAL(18,2) DECIMAL(18,5) DECIMAL(18,6) See: Note 1 and Note 6	-	+	+
Upsolver	BOOLEAN STRING Number$^{6}$ See: Note 6	-	+	+
Redshift	N/A See: Note 3	-	+	+
MySQL	N/A See: Note 3	+ See: Note 4	+	+
Apache Kafka	N/A See: Note 3	-	+	+
Amazon Kinesis	N/A See: Note 3	-	+	+
Elasticsearch	N/A See: Note 3	-	+	+
Amazon S3	N/A See: Note 2		+	+
HDFS	N/A See: Note 2		+	+
Google Storage	N/A See: Note 2		+	+
Microsoft Azure Storage	N/A See: Note 2		+	+
Qubole	DOUBLE$^{6}$ BIGINT$^{6}$ BOOLEAN STRING TIMESTAMP DECIMAL(18,2) DECIMAL(18,5) DECIMAL(18,6) See: Note 6	+	+	+
Amazon SageMaker	N/A See: Note 2	-	+	+
Lookup Table	BOOLEAN STRING NUMBER TIMESTAMP NESTED TYPES	+	+ See: Note 5	+

Note 1

Not fully automated.

Note 2

The type is selected based on the output types. You can cast fields to get any types supported by the output format.

Note 3

You cannot choose the destination data type as it is fetched from the original database.

Note 4

Supported by definition. If a Primary Key (PK) is configured in the database, we will only have the latest record per PK.

Note 5

Only aggregations by key are supported for lookup tables.

Note 6

DOUBLE, BIGINT and Upsolver NUMBER data types are limited to 64 bits.