More and more companies realise the dependency on the availability of the network and the internet. As enterprises expect 100% percent uptime of the network and demand always-on internet and email has become the nummer 1 business application, network engineers need to redesign the network and eliminate single points of failures Single points of failures are those elements in the network that if they fail, it will result in serious loss of connectivity.

Typical single points of failure:

• (core) network switch
• firewall
• internet connection
• server network interface
• router

Eliminate Single Points of Failure

In our (SecureLayers) opinion, elimination of single points of failures should be part of business contingency planning. Elimination of the single points of failures can be implemented fairly quickly and with minimal impact. Depending on the business needs, the amount of downtime it can sustain and the available budget, a network redesign will be drawn up and the necessary (fixed price) budget.

Also low-cost solutions can be implemented to eliminate single points of failures, like the use of a consumer DSL line or a prepaid 3G connection to provide a fail over connection in case the primairy internet connection is down.

Monitoring

When the network is more robust and the redundant elements are implemented, one should not forget to implement network monitoring. Because in a full redundant network a single failure will not lead to network downtime or loss of functionality, so all the elements should be monitored so mail function can be indentified and remedied.