Replication is the process of copying/moving data between 
databases on the same or different servers. SQL Server 
supports the following types of replication scenarios: 

Snapshot replication 
Transactional replication (with immediate updating 
subscribers, with queued updating subscribers) 
Merge replication 

Replication is way of keeping data synchronized in multiple 
databases.SQL server
replication has two important aspects publisher and 
subscriber.
Publisher:
Database server that makes data available for replication 
is called as Publisher.
Subscriber:
Database Servers that get data from the publishers is 
called as Subscribers.
There are three types of replication supported by SQL 
SERVER:-
Snapshot Replication.
Snapshot Replication takes snapshot of one database and 
moves it to the other
database.After initial load data can be refreshed 
periodically.The only disadvantage of
this type of replication is that all data has to copied 
each time the table is refreshed.
Transactional Replication
In transactional replication data is copied first time as 
in snapshot replication , but later
only the transactions are synchronized rather than 
replicating the whole database.You
can either specify to run continuously or on periodic basis.
Merge Replication.
Merge replication combines data from multiple sources into 
a single central database.Again
as usual the initial load is like snapshot but later it 
allows change of data both on subscriber
and publisher , later when they come on-line it detects and 
combines them and updates accordingly

Replication is a method of copying data from one database 
to other database or one server to another 

there are three types of replication 

(1) transactional (2)snapshot (3) merge (4) logshipping and 
in 2005 one more type is introducde which was called as 
peer to peer replications

Replication is the technique provided by MS SQl server to 
keep the replicate copy of database.

There are four basis type of Replication which we can set 
up in sql server

1. Snapshot replication – It is used when data at Publisher 
is static means not changing frequently and Subscriber need 
update data

2. Transactional replication - It is used when data is 
changing frequently at publisher and Subscriber need on 
line data from publisher.

3. Merge replication- It is used when there is no constant 
connection between Publisher and Subscriber. When they come 
on line then data will       replicate. In replication data 
can conflict and resolve this there is rule set up already.

4. Transactional replication with Updateable Subscription-
   It is similer to Transactional replication but update 
backs to Publisher from Subscriber. In this it add extra 
field in table called ms_repl- etc to keep the update data 
from Subscriber.


  
We can also implement Peer to Peer replication which is 
Headless topology. To do this we need enterprise edition of 
sql server 2005. In this there is more than one Publishers 
and one or more Subscribers. Every participant is both 
Publisher and Subscriber. This is mostly use when every one 
wants to read and modify the data.