Disk Storage Replication
NAS Recovery - Disk Storage Replication
Disk storage replication is the process of sharing information by distributing updates of a block device to several physical hard disks. Storage replication is implemented either in a disk array controller (hardware) or a device driver (software). This method of replication works in any file system supported by the operating system because the file system code works on a level above the block device driver layer.
The basic form of disk storage replication is disk mirroring. It is implemented in locally connected disks. A replication, on the other hand, is extendable across a computer network where the disks are placed in physically distant locations.
Disk storage replication is useful because it improves the availability of the system even in case of local failures or disasters.
There are different types of replication methods:
Synchronous replication: This method uses an ‘atomic’ write operation in which the write process is completed on both sides or none. The write operation is considered complete only after it is acknowledged by both local and remote storage. This process ensures zero data loss. However, it may slow down the overall performance of the network because most applications wait for a write transaction to complete before proceeding with further work.
Asynchronous replication: In this method, the write process is considered complete as soon as local storage acknowledges it. Remote storage will be updated with a small lag. This method ensures greater performance, but there is always a risk of data loss. In case of a data loss at the local storage, the remote storage may not have the current copy of data and most recent data may be lost.
Some of the popular implementations include: DRBD module for Linux, IBM Copy Services, Hitachi TrueCopy and Symantec Veritas Volume Replicator.
