Introduction › Supported Application and Database Servers › How Synchronization Works › Synchronization Methods

Synchronization Methods

To properly synchronize the Master and the Replica, it is first necessary to perform a comparison of their two file structures. This comparison determines what content (files and folders) on the Master is missing or different from the content on the Replica. You have two ways to start Synchronization:

• Click the Synchronize button on the Manager toolbar
• Click the Run button on the Manager toolbar

Select a synchronization mode, each with a different comparison algorithm and method of operation:

File synchronization

In file synchronization, the Replica server sends its snapshot to the Master server, which uses it to drive the information and content for the data comparison. After the comparison is performed, the Engine on the Master server sends a sequence of commands to the Replica. These commands:

• Delete files that exist only on the target
• List the entire file contents of files that either exist only on the Master, or that exist on the Replica but differ from the version on the Master

This method is best suited for File Servers or application servers with many relatively small files.

Note: When performing File Synchronization, Arcserve RHA does not update the data transferred percentage until the file transfer is completed.

Block synchronization

In block synchronization, the Engine performs a block-by-block comparison of the Master and Replica files, and copies over only those blocks that are different. When differences exist between files, instead of requiring the transfer of the entire file, block synchronization transfers changes only.

This method is best suited for database applications, such as Microsoft Exchange Server, Oracle, and SQL Server, or application servers with large files.

Offline synchronization (Available only from the Run dialog)

In offline synchronization, data is copied from the Master server to an external device, and from the external device to the Replica server. Full System scenario supports Offline synchronization method for forward, backward, and BMR restore scenario.

This method is best suited for transferring huge data volumes without much impact from a low network bandwidth. This option is available only when running a scenario and does not apply to scenarios with scheduled replication, or scenarios in which the Master is a UNIX/Linux host. For more information, see How Offline Synchronization Works.

Volume Synchronization (Available only for Full System Scenarios)

In volume synchronization, the Master server copies the volume as whole instead of a file or a block. All volumes on a disk are synchronized at a volume level. RHA accesses the bitmap and only copies the used data. For example, when the used volume is 25 GB of a 40 GB volume, only the used 25 GB is copied.

Note: If the size of volumes on the Master is larger than the size on Replica, RHA switches to block synchronization.

The comparison can be configured to consider only file size and modification time to determine whether two files differ, or it can perform a check of the actual contents of the data. The former approach, which is not valid in the case of database applications, can be a legitimate way to increase the comparison process on a File Server scenario.

You can Filter or Skip synchronization.