Today's automotive software designs increasingly target high-capacity electronic control units (ECUs) that incorporate extremely powerful embedded microcontroller units (MCUs). This trend enables connected and autonomous vehicle technologies to move forward. Modern MCUs have been developed with multiple cores to support this shift to software-defined vehicles, enabling parallel execution of vital software functions.
This white paper discusses the challenges software teams face when designing an architecture to accommodate these high-capacity ECUs, focusing on application and communication interactions, and examining best practices to improve overall performance.
Today's vehicles contain an interconnection of ECUs. As automotive functionality increases, certification and compliance become increasingly more difficult. One approach is to continue adding multiple ECUs to a vehicle's embedded system and to rely on the merits of these and multicore processing. Another theory is to condense multiple ECUs into a single onboard supercomputer. Multicore processing is the way to go if performed correctly.
The problem to be addressed by the automotive industry is the parallel execution of a parallel application— How many applications can be localized to the same ECU so independent vehicle functions can be executed in parallel? Dependent functions can often be partitioned so that they are heavily parallelizable. This paper will highlight a few possible scenarios.
Siemens AUTOSAR solution, Capital™ VSTAR software, is the natural choice for the embedded system running on a multicore MCU to deploy high-power processes that require different dependencies to resources while maintaining focus on application and integration. Developers are aided by a generative workflow that creates a correct-by-construction embedded software platform allowing the engineers to focus on innovative application development.
Download the white paper to learn why multicore processing is the way forward – and for the right reasons.