Optimal Multicore Processing for Safety-Critical Applications

Optimal Multicore Processing for Safety-Critical Applications
Authored by Green Hills Software featuring Curtiss-Wright Defense Solutions

Safety-critical and Multi-Level-Security (MLS) applications have been slow to utilize multicore architectures due to the complexity of validating and certifying software and hardware architectures. Of principal concern is how an application running on one core can interfere with an application running on another core, thereby affecting the determinism, quality of service, and ultimately safety. Yet, if the concerns over multicore operation can be addressed, the benefits of smaller size, lower power, and increased performance can be realized.

Optimal Multicore Processing for Safety-Critical Applications

Figure 1: Separate processor cores (gray) share many resources (green) ranging from the interconnect to memory and I/O.

In order to achieve the throughput and SWaP benefits of multicore solutions, the software architecture needs to support the high utilization of the available processor cores. This requires support for full multicore features, ranging from enabling concurrent operation of cores, instead of available cores being forced into an idle state or held in reset at startup, to provide a mechanism for deterministic load balancing. In general, the more flexible the software multiprocessing architecture is, the more tools the system architect has to achieve high utilization.

Download the white paper to learn more.

  • The benefits of multicore processors, such as higher throughput, better SWaP, room for future growth, and longer supply availability
  • Challenges for multicore processors in safety-critical applications, and how to overcome them
  • Software multi-processing architectures and essential multicore RTOS requirements for critical airborne systems