That's a great question. We take security and reliability very seriously here at Screenly. All of our critical services run in clusters with full redundancy and automatic failover between nodes. Thus, we have designed our entire infrastructure, including all network services and production servers, from the ground up with fault tolerance in place and no single points of failure. We rely on a number of different databases that hold sensitive customer data, and also host offsite backups that cannot be tampered with. If we were to get compromised (and/or a dataset were to get corrupted), we have multiple levels of copies that we would be able to restore from.