Microsoft SQL Server Source and a Target That Uses the Parquet Output Format

The following table identifies the recommended data-type mappings for Database Ingestion and Replication configurations with a Microsoft SQL Server source and an Amazon S3, Google Cloud Storage, Microsoft Azure Data Lake Storage Gen2, or Microsoft Fabric OneLake target that uses the Parquet output format:

Microsoft SQL Server Source Data Type	Parquet Data Type
bigint	init64
binary	byte array
bit	int32
char	string
date	date
datetime	timestamp
datetime2	timestamp
datetimeoffset	string
decimal(p,s)	decimal(p,s)
float	string
geography	byte array
geometry	byte array
hierarchyid	byte array
image	byte array
int	int32
money	decimal(19,4)
nchar	string
ntext	string
numeric(p,s)	decimal(p,s)
nvarchar	string
real	string
smalldatetime	timestamp
smallint	int32
smallmoney	decimal(10,4)
sql_variant	byte array (initial load) string (incremental load)
text	string
time	string
timestamp	byte array
tinyint	int32
uniqueidentifier	string
varbinary	byte array
varchar	string
xml	string

Note:

You can't override these mappings by defining data-type rules on the Task Details - Target Details page in the task configuration wizard. If you need custom mappings, contact Informatica Global Customer Support.