System scalability starts with the operating system. QNX Neutrino, by nature of its very architecture enables systems to scale from tiny embedded devices to some of the highest capacity networking equipment available today.

QNX Transparent Distributed Processing (TDP) provides a highly distributed loosely coupled multiprocessing capability that is independent of the underlying interconnect technology. TDP allows resources to be shared between processor cores in a single chip, between processors on a board and even between line cards over any backplane technology.

Increasingly, processor manufacturers are turning to multi-core implementations to scale performance. While this addresses the hardware requirement for increased MIPS, it does not address the burden now placed on the software developer to migrate existing code onto this new architecture.

QNX offers the world’s most comprehensive support for multi-core processors. QNX Neutrino is the only RTOS that offers proven, deployed symmetric multiprocessing capability. Where other OS vendors are struggling to address the new multi-core model, QNX customers have benefited from QNX multi-processing technology for over a decade. QNX for multi-core processors supports both the asymmetric and symmetric multiprocessing models. QNX has gone even further, pioneering bound multi-processing, a technology that spans the asymmetric and symmetric processing models to offer the control and determinism of asymmetric multiprocessing with the simplified programming model and performance benefits of symmetric multiprocessing. QNX Momentics was developed from the ground up with multiprocessing capabilities. The combination of an instrumented kernel and advanced visualization tools provide developers with the ability to quickly and effectively debug and optimize code in a multi-core environment.

To address the need to scale development teams across multiple subsystems, QNX provides Adaptive Partitioning. With Adaptive Partitioning, system architects can allocate CPU and memory resources to each development team. Development teams are then able to complete their subsystems within their budgets. At system integration, complexities arising from resource contention and conflicting priorities are eliminated. Unlike traditional partitioning schemes, the dynamic nature of Adaptive Partitioning ensures maximum CPU utilization by ensuring that unused CPU cycles from one partition are made available to other partitions should they be able to make use of them.