Pages

What is Failover and Clustering ?

First we go for the definition of Failover.
Failover is the capability to switch over automatically to a redundant or standby computer server, system, or network upon the failure or abnormal termination of the previously active application,server, system, or network.

source : http://en.wikipedia.org/wiki/Failover
In simply failover is something like this. When client request something from the server but the particular server is fail the request automatically goes to another server and respond is goes to the client. But in backend the main server is check when the fail server is recovering. After the server is recover the client request automatically redirect to that server.

The main advantage is availability and the high degree of reliability.

There are different types of failovers.

source : http://knol.google.com/k/failover

There are some failovers those are not automatically it require some manual intervention.
In Clod failover must do in manually.
Warm failover means it transact to other server automatically but the current transaction may abort because of the data synchronization failure.
Hot failover is the most reliable failover. Because it transact to another server in automatically and 100% synchronize the data. So there is no failure between the client and the sever.

In practically failover is very important to all of us. Suppose if we request a streaming video, streaming audio,VoIP or downloading a file from the internet. If the particular server is fail the connection between the server and the client is abort. So all the operation is abort. But if there is a failover capability it will automatically redirect the request to another server and continue the operation.  
So that's all about failover.

Now we will look into Clustering.
Here is the definition.
A computer cluster is a group of linked computers, working together closely thus in many respects forming a single computer. The components of a cluster are commonly, but not always, connected to each other through fast local area networks.

source : http://en.wikipedia.org/wiki/Computer_cluster

Clustering means group of computers word together to provide some enterprise service. 
Clusters can comprise the redundant and failover capable.

All clusters provide two main advantages.
Scalability and High Availability.
Scalability means application support increasing number of users. Because clusters provide extra capacity by adding new servers.


Cluster Categorization
High Availability Clusters
Load Balancing Clusters
Compute Clusters

There are two types two Clusters
1.Shared nothing
Every application server has a files. Updating and maintenance files is very hard. 
2.Shared disk
There is storage disk that contains the files and application servers use those files. So update and maintenance files is easy.

No comments:

Post a Comment