DB2 Datatype Compatibility

Replication Agent for UDB processes transactions and passes data to the primary Replication Server. The primary Replication Server uses the datatype formats specified in the replication definition to receive the data from Replication Agent for UDB.

This table describes the default conversion of DB2 datatypes to Sybase datatypes. For each datatype in this table, lengths in the second column are described as:

  • Character datatypes – maximum number of bytes.

  • Graphic datatypes – maximum number of characters.

  • Numeric datatypes – range from smallest to largest values.

  • Temporal datatypes – range from earliest time to latest time.

DB2 to Sybase Default Datatype Mapping

DB2 Datatype

DB2 Length/Range

Sybase Datatype

Sybase Length/Range

Notes

BIGINT

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

bigint

10-38 to 1038, 38 significant digits

BLOB

Variable-length, 2GB, binary data

image

2GB

CHAR

254 bytes

char

32K

CHAR FOR BIT DATA

254 bytes, binary data

binary

32K

CLOB

Variable-length, 2GB, character data

text

2GB

DATE

0001-01-01 to 9999-12-31

char, date, or datetime

32K (char)

Use the Replication Server heterogeneous datatype support (HDS) feature for datatype conversion and translation.

DBCLOB

Variable-length, 2GB, double-byte character data

unitext or image

2GB

For Replication Server 15.0 and later versions, DBCLOB maps to unitext. For earlier versions of Replication Server, DBCLOB maps to image.

DECFLOAT(16)

8 bytes, -9.999999999999999 x 10384 to -1.0 x 10-383 and1.0 x 10-383 to 9.999999999999999 x 10384

float

Precision and range corresponds to a C double datatype, approximately 16 significant digits

Some loss of precision may result.

DECFLOAT(34)

16 bytes, -9.9999999999999999 99999999999999999 x 106144 to -1.0 x 10-6143

and

1.0 x 10-6143 to 9.999999999999999999999999999999999 x 106144

float

Precision and range corresponds to a C double datatype, approximately 16 significant digits

Some loss of precision may result.

DECIMAL

-1031+1 to 1031-1, 31 digits of precision

decimal

10-38 to 1038, 38 significant digits

  

DOUBLE

See FLOAT.

FLOAT

8 bytes, -1.79769308 to 1.79769308

float

Precision and range corresponds to a C double datatype, approximately 16 significant digits

Extremely small values are truncated to 16 digits to the right of the decimal. Extremely large values retain their precision.

GRAPHIC

127 characters, double-byte character data

unichar

32K

INTEGER

-2,147,483,648 to 2,147,483,647

int

-2,147,483,648 to 2,147,483,647

LONG VARCHAR

Variable-length, 32,700 bytes, character data

text

2GB

LONG VARCHAR FOR BIT DATA

32,700 bytes, binary data

image

2GB

LONG VARGRAPHIC

16,350 characters, double-byte character data

unitext or image

2GB

For Replication Server 15.0 and later versions, LONG VARGRAPHIC maps to unitext. For earlier versions of Replication Server, LONG VARGRAPHIC maps to image.

NUMERIC (synonym for DECIMAL)

See DECIMAL.

REAL

-3.40238 to 3.40238

decimal

10-38 to 1038, 38 significant digits

SMALLINT

-32,768 to 32,767

smallint

-32,768 to 32,767

TIME

00:00:00 to 24:00:00

char, time, or datetime

32K (char)

TIMESTAMP

0001-01-01-00.00.00.000000 to 9999-12-31-24.00.00.000000

char or datetime

32K (char)

Use the Replication Server heterogeneous datatype support (HDS) feature for datatype conversion and translation.

VARCHAR

32,672 bytes

varchar

32K

VARCHAR FOR BIT DATA

32,672 bytes, binary data

varbinary

32K

VARGRAPHIC

16,336 characters, double-byte character data

univarchar

32K

Parent topic: IBM DB2-Specific Considerations