The platform is designed in a way that even in the unlikely event when all of its control servers go offline, the active Cloud servers on the hypervisors will continue to process uninterrupted.
The redundancy is organized in an N+1 configuration, wherein the content and configuration of the main server is replicated in real time on at least one other server, which can automatically replace the first in case of failure.
A broad range of the server parameters are monitored 24x7x365 in real time through a special IPMI controller and a specialized monitoring system - temperature of the processors of the whole system and hard drives, performance of the cooling fans, voltage values of the power modules, S.M.A.R.T status of the hard disk drives, used disk space, RAM load, processor load, network connections, number of running processes, etc. The monitoring allows on-time diagnosis, prevention or troubleshooting of possible problems on the server.
The network connectivity of the servers is completely redundant and all connections go through several independent network controllers and devices.