Fault tolerance is the heart of what a NonStop is. The high availability of NonStop systems comes from close cooperation between the hardware and software. Every component in an HPE NonStop server is duplicated. If one component fails, there is instantaneous failover to a backup. This includes not only hardware components, such as processors, memory, I/O channels, and power supplies; but also all software components, such as system and application processes. As a consequence, a NonStop server is self-healing in the presence of any single fault and many multiple faults. If a component fails, no one notices except for the IT staff that must correct the fault.
Scalability is important to me. I might need a performance boost at certain times of the year, especially around the Christmas period. In a company I previously worked for, we payed particular attention to the number of POS and ATM transactions around the Christmas period because this was known to be the busiest period of the year. It would have been nice not to have to worry about the transaction rate and CPU/DISK performance, but to simply turn on a CPU core for that particular time of year. Since upgrading the hardware, this worry is a thing of the past.
Security is always important, and the NonStop is no different. Many security layers make it very difficult for hackers to access information you don’t own or don’t need to see, even if you know what you are doing.
There is never any downtime due to the way the hardware and software interact. They are both fault tolerant and, therefore, give a robust service.
