An Introduction to Database Replication
The volumes of data generated by businesses today are massive. IT departments are expected to ensure that this data is online and accessible indefinitely, thereby putting immense pressure on the data management process. The need then is to replace outdated and inefficient legacy processes through agile and new techniques. SQL Server Replication is the most used technique to achieve this goal.
But first – what is SQL Server Replication? It is a technology for copying and distributing objects from one database to another after which the databases are synchronized to maintain integrity and consistency of data. Database replication SQL server is used for continuous copying and synchronizing data or scheduled to run at intervals that are determined from time to time. The data synchronization can be carried out in several ways as per the replication techniques deployed.
Here are some of the techniques used for data replication.
Snapshot replication – In this process, a “snapshot” is taken of data on one server which is then moved to another server or sometimes another database on the same server. After the first synchronization snapshot, replication will refresh data in published tables periodically depending on what is preset. This technique might seem quite easy to set up and maintain but all data has to be copied each time a table is refreshed. Between refreshes, data on the publisher and subscriber might differ.
Transactional replication – It is the process of copying data from the publisher once to the subscriber and then delivering transactions to the subscriber (or more than one subscriber) as they happen on the publisher. The first copy of the data is transmitted through the same mechanism as with snapshot replication. As and when the publisher inserts updates or deletes records, the transactions are forwarded to the subscriber(s).
To ensure quick database replication SQL server, a configuration command may be generated to deliver transactions continuously. You can even run the synchronization tasks periodically. This form of replication works best where there is a dedicated and dependable line between database servers taking part in the replication process. Typically, database servers that subscribe to transactional publications use data strictly for read-only purposes without scope for modifications.
Merge Replication
In this replication technique, data from multiple sources are combined in a single database. As in transactional replication, initial synchronization is done by taking a snapshot of data on the publisher and moving it to subscribers. However, unlike transactional replication, merge replication allows changes to be carried out on the same data on publishers and subscribers even though subscribers may not be connected to the network. Once subscribers connect to the network, this process will change data on the publisher by detecting and combining changes from all subscribers. Merge replication is useful when there is a need to modify data on remote computers and subscribers are not assured of a stable and guaranteed link to the network.
This in a nutshell is what database replication SQL server is all about.
But first – what is SQL Server Replication? It is a technology for copying and distributing objects from one database to another after which the databases are synchronized to maintain integrity and consistency of data. Database replication SQL server is used for continuous copying and synchronizing data or scheduled to run at intervals that are determined from time to time. The data synchronization can be carried out in several ways as per the replication techniques deployed.
Here are some of the techniques used for data replication.
Transactional replication – It is the process of copying data from the publisher once to the subscriber and then delivering transactions to the subscriber (or more than one subscriber) as they happen on the publisher. The first copy of the data is transmitted through the same mechanism as with snapshot replication. As and when the publisher inserts updates or deletes records, the transactions are forwarded to the subscriber(s).
To ensure quick database replication SQL server, a configuration command may be generated to deliver transactions continuously. You can even run the synchronization tasks periodically. This form of replication works best where there is a dedicated and dependable line between database servers taking part in the replication process. Typically, database servers that subscribe to transactional publications use data strictly for read-only purposes without scope for modifications.
Merge Replication
In this replication technique, data from multiple sources are combined in a single database. As in transactional replication, initial synchronization is done by taking a snapshot of data on the publisher and moving it to subscribers. However, unlike transactional replication, merge replication allows changes to be carried out on the same data on publishers and subscribers even though subscribers may not be connected to the network. Once subscribers connect to the network, this process will change data on the publisher by detecting and combining changes from all subscribers. Merge replication is useful when there is a need to modify data on remote computers and subscribers are not assured of a stable and guaranteed link to the network.
This in a nutshell is what database replication SQL server is all about.
Comments
Post a Comment