Nuclias Cloud has a 99.9% uptime SLA level which is equates to a permissible 43 minutes and 49 seconds downtime in a month. Unlike traditional software controller solutions pre-installed onto appliances or a dedicated server, which create a single point of failure as well as performance and scalability challenges, Nuclias Cloud’s design spreads its capabilities and load balances over different back-end clustered servers.

These clustered servers are virtually provisioned, so Nuclias Cloud can scale its back-end infinitely as demand increases. By removing local control, Nuclias decentralises the network architecture and eliminates the single point of failure which you normally experience in traditional network architecture.  Any disruption (power cut, loss of network connectivity or even hardware failure) in an area of the network has no impact on the remaining network. 

For fault tolerance purposes, Nuclias Cloud has automatically backs-up clusters running in different data centers so that these clustered servers can be activated in less than 10 minutes if something unforeseeable happens. 

Also, all Nuclias Cloud devices are designed to operate in standalone mode, if connectivity to the Nuclias Cloud is lost, ensuring that connectivity for the end-user is not impact on either the wired or wireless LAN.  In the event that connectivity is lost, then the only impact is that log data exchange between the cloud and the device is suspended until the connection is restored.  Since this data exchange is generally log and or system update/configuration changes from the cloud to the device a delay has no impact on the performance of the network. 
Nuclias cloud provides flawless controllability as well as cloud availability.