No system can ever be 100% reliable, but our servers use the best practice in server technology to minimise any downtime:
If something fails, another system will usually kick in automatically to keep the service running.
If a problem occurs that requires human intervention, Engineers are on standby and almost always have the issue resolved before our customers realise that anything has gone wrong.