Using extended limits

You can create replication definitions using extended limits for both replicate and standby databases. Extended limits are defined as:


Wide columns

Replication Server can replicate wide columns containing char, varchar, binary, univarchar, unichar, unitext or Java inrow data to a maximum of 32768 bytes. Maximum column width on each system may vary; it is a function of the total number of columns and the page size of the data server.

You can use wide columns as primary keys and searchable columns and in replication definition where clauses.

NoteThe maximum number of bytes in the where clause of a subscription or article is 255 bytes. You cannot use wide columns in the where clause of subscriptions or articles.


More columns

You can include as many as 1024 columns in a replication definition. As long as the total number of columns does not exceed 1024, Replication Server does not limit the number of primary key or searchable columns.

Replication Server uses primary key columns to build where clauses of SQL statements for the data server. Consider data server limitations when determining the actual number of columns available for primary keys in replication definitions.

Similarly, although Replication Server imposes no limits on the n umber of searchable columns in a replication definition, the number of columns in the where clause of a subscription or article may also be constrained by data server limitations.


Wide data

Data rows can equal the size of the data page on the data server. Adaptive Server version 12.5 and later supports page sizes of 2K, 4K, 8K, and 16K.


Wide messages

Replication Server copies data rows as messages in stable queues manage by the SQM. These messages contain before and after images of replicated data rows as well as other information. They require significantly more space than the data rows on which they are based. With extended limits, messages can span blocks and are no longer limited to 16K.