Replication is the process of sharing any level of information so as to ensure consistency between redundant hardware/software resources to improve reliability, fault-tolerance, and accessibility
The concept of replication have been performed either asychronously/asychronously for DB and NFS Servers.
From the perspective of databases, this has been performed asychronously via log shipping: the copying of log files containing completed SQL transactions to be performed on one or more database servers setup as Slaves. MySQL, PostgreSQL, and SQL Server support this type of replication. MySQL has improved upon this in MySQL 5.5 with semisynchronous replication. PostgreSQL has improved upon this with Streaming Replication.
From the perspective of OS Disks, this has been accomplished by network RAID1 schemes such as DRBD. DRBD performs block-level synchronous replication betweent two disks over a network (preferably private netblocks). This type of replication does not perform well over geographic distances. On the other hand, a DRBD Server pair should be using crossover cabling on a local subnet for the fastest possible disk replication by shortening the network transmission distance. The use of DRBD can provide an additional layer of redundancy for products such as MySQL, PostgreSQL, SQLLite, and Ingres.
NoSQL databases, such as Cassandra and MongoDB, also perform replication. Excellent replication performance can be achieved with proper tuning and and enhancement of server networks.
