There is often confusion as to when to use Sync Frameworks or Replication. Sync Frameworks can be used to synchronize data stores whether they are databases or even a file store whereas replication is about synchronizing SQL databases.
These two synchronization mechanisms also target different audiences. Sync Frameworks targets the developers and provides more flexibility. It also imposes more coding as in the case of the non-database store, the developer must code the providers. Replication targets the DBAs and furnishes out of box procedures and monitors.
The following table also provides insight into how each mechanism addresses key features:
|
Key feature |
Merge replication |
Sync Services |
|
Synchronize by using an n-tier or service-oriented architecture |
No |
Yes |
|
Supports heterogeneous databases |
No |
Yes |
|
Incremental change tracking |
Yes |
Yes |
|
Conflict detection and resolution |
Yes |
Yes |
|
Easily create data views on the client |
No |
Yes |
|
Automatically initialize schema and data |
Yes |
Yes |
|
Supports large data sets |
Yes |
Yes |
|
Query processor is locally available |
Yes |
Yes |
|
Automatically propagate schema changes |
Yes |
No |
|
Automatically repartition data |
Yes |
No |
More info can be found at: