Highly available equipment is a cornerstone in any networking or telecom design. QNX Neutrino offers unparalleled availability both by virtue of its memory protected microkernel architecture and its high availability middleware.

QNX Neutrino offers far greater fault isolation and dynamic upgradeability than any other offering available. Not limited to application level fault isolation, QNX Neutrino extends fault isolation to all elements in the system. Because everything outside the micro-kernel exists in memory protected user space, this protection exists not only for applications but also includes file systems, protocol stacks and even device drivers. In this model, faults are limited to the process in which they occur — device drivers, filesystems, stacks and applications are all processes in the micro-kernel model). Faults are limited to the process in which they occur and bugs that might otherwise take weeks to track down are identified immediately.

Availability is a combination of mean time between failures and mean time to repair. The memory protected micro-kernel model offers significant advantages in both aspects. By providing memory protection to all aspects of the system, mean time between failures improves dramatically. The inherent modularity of QNX Neutrino also improves mean time to repair. Processes and therefore patches and upgrades can dynamically deployed to in-service systems without having to reboot the device. With QNX Neutrino, high availability is a function of the operating system architecture and not a bolt-on afterthought.

To supplement the high availability inherent to the QNX architecture, QNX offers a high availability framework. This framework enables developers to create custom recovery scenarios. A smart watchdog monitors process status via heartbeating. In the event of a process failure, user defined recover scenarios are executed.

Applications, stacks and even drivers can be intelligently restarted without having to reboot the system.

QNX Adaptive Partitioning provides resource guarantees to defined partitions, ensuring that all elements of the system enjoy guaranteed access to critical resources. Examples of how Adaptive Partitioning can increase availability include containment of DoS attacks and preventing the runaway CLI.