In addition, both the master and the mirror report independently and simultaneously to other critical software components. These processes can be mirrored and spread across the carrier network across multiple data centers such that no single "box" can halt the software from working properly. This means all of the active state tables, statistics, load balancing and other real-time data are always preserved along with calls in progress. Even a single hour of downtime can have severe financial consequences - making hot-backup essential to address contingent liabilities. __ Hot Back-Up, Not Fail-Over or Hot-Standby: This mirrored architecture offers true ''Hot backup" of all CallCenter@nywhere software processes. In practice, if the (hardware) server that is running a software process crashes, another server on the network can instantly take over without the loss of any information or the disconnection of calls - even on calls in progress. This Hot Backup is distinguished from traditional "Fail-Over" schemes. With "Fail-Over", calls and all other forms of interactions are disconnected in the event of system failure -- and then a second system takes over as of the next call. With "Fail-Over, thousands of calls may have to be disconnected and the back-up system requires those callers to call back and has to start over their calls over again. The distinction is significant to companies, who require mission-critical communications infrastructures. The ability to stay up and running all of the time when you're processing thousands of transactions has been daunting up until now. You can have all possible hardware redundancy but without software that mirrors itself and its active processes on the network, the best that you can hope for is that your customers will forgive the interruption of service. Now, processes can be distributed on more than one physical server across multiple locations so that actual live transactions and their associated data remain safe at all times.