| 15 October 2009
Acluster of high availability (Linux HAS) is a set of two or more machines, which are characterized because the failure in one of the machines does not make stop the service that they offer as a whole. If there takes place a failure of the hardware or of the applications marked like critical of someone of the machines of the cluster, our system of monitoring of high availability is capable of re-starting automatically any of other machines of the cluster.
A cluster is hoped to present combinations of the following services:
- High performance
- High availability
- Balance of load
- Scalability
The aim of design of the servants of High Occentus Network's availability is of providing availability and reliability. These clusters try to offer the maximum availability of the services that offer. The reliability is provided by means of software that detects failures and allows to recover opposite to the same ones, whereas in hardware it is avoided to have the only point of failures.
In the basic unfolding one proposes the employment of 2 identical servants under an ip address Failover RIPE. This is the dynamic management of the traffic before falls of the service. The service is given by a principal servant (A) and by means of copy DRBD (remote copy of the hard disk) in the servant of backup (B).
The housing of the servants takes place in centers of information that rely on certification ISO27001:2005 in safety and management of the information, and direct connection to ESpanix and CATnix, optimizing the connection and stable latency in the whole Spain.
Construct your own net of high availability with Occentus Network
Construct your servants' own cluster. Select two or more dedicated (identical) servants, together with the service of representative management. These will be installed together depending physically on the same Switch. This cluster can be formed in way High-Availability (HAS) or Load Balancing (LB).
