After the initial full synchronization is complete, vSphere Replication starts to track the changed blocks on the source site and periodically transfers them to the target site. This process is called an incremental sync. As a result of the incremental sync completion, vSphere Replication creates a new replica instance on the target site. The following removal of the old instance is a time-consuming process.

During the old instance removal process, vSphere Replication might start transferring new changed blocks to the target site. This activity further increases the storage consumption on the target site. When the old instance removal is complete, vSphere Replication frees the storage space occupied by the old instance. If the source disk has a high data change rate, while the old replica instance is being removed, the storage consumption on the target site might temporarily exceed several times the size of the source disk.

Note: If due to the temporary spikes in the storage consumption on the target site the space is not enough, you might observe "Insufficient storage space" errors in the Site Recovery user interface. vSphere Replication might start reporting recovery point objective (RPO) violations.