Exchange 2007 Replication Introduction

The original RTM release of Exchange Server 2007 offered two different types of continuous replication: local continuous replication and cluster continuous replication. Local continuous replication provides a degree of fault tolerance on a single mailbox server. The basic premise is that LCR creates a secondary copy of an Exchange Server database on a separate volume than the server. In contrast, CCR uses a similar method to create a separate copy of the database and stores it on a separate Exchange server.

When Microsoft released Service Pack 1 for Exchange Server 2007, it introduced a third type of continuous replication known as standby continuous replication (SCR). SCR is similar to CCR; however, CCR can only create one replica of a protected database. SCR can create multiple replicas of a database.

Even though Exchange Server 2007 provides three different types of continuous replication, all three work in basically the same way. Continuous replication is based on a technique called log shipping, which involves copying each log file to the location where the backup database is stored while the log file is built. The log file is then replayed against the replica database, bringing the database into a current state.

Microsoft classifies continuous replication as a storage group-level operation, but I think it's more accurate to classify it as a database-level operation. Log files reside at the storage group level, but continuous replication, in any form, limits the protected storage group to a single database. Therefore, continuous replication could be considered a database-level operation because it involves a single database.