Planning for Each Type of Replication
Each type of replication (snapshot replication, transactional replication, and merge replication) has specific requirements and issues that you should consider before implementation.
Because an initial snapshot must be applied for all types of replication, you should be familiar with the planning considerations for snapshot replication even if you choose to implement transactional replication or merge replication.
When considering transactional replication, allocate adequate disk space in the distribution database to handle the number of transactions that will be stored there.
When considering merge replication, Microsoft® SQL Server™ 2000 uses a globally unique identifier (GUID) column to identify each row during the merge replication process. If the table that is replicated does not have a uniqueidentifier column with the ROWGUIDCOL property and a unique index, SQL Server 2000 will add one to the table, and you will need to account for the additional data that is stored there. If the table already has a uniqueidentifier column, you can add the ROWGUIDCOL property to signal that it can be used during merge replication. You must also add a unique index on this column or make it the primary key for the table. Distributed applications can benefit greatly from using the uniqueidentifier column because it guarantees that no two sites will generate the same key value.