By Marques McCammon
Work smarter, not harder. OEMs will increasingly need to do more with less. In any businesses model, but particularly in the auto industry, there is a constant focus on materials cost to preserve margins all while their products become more technologically sophisticated.
This brings us to the next call to action in Wind River’s software strategy ACRU model, “CONSOLIDATE.” As mentioned in my previous post, the ACRU model is a strategy designed to help car makers keep pace with consumer demands and upgrade legacy technology without constantly adding more cost, weight, power consumption, and complexity that eats away at profits.
Over the years, OEMs have added more ECUs inside the car as they create new capabilities and types of experiences – a very linear and highly costly direction. The challenge is that additional ECUs can typically require more power drain and additional weight due to new materials used. How can this be mitigated? OEMs must try to consolidate these new compute workloads within the car as much as possible.
While many automakers and tier-one suppliers are already investigating consolidation, there is more that can be done. Wind River sees an opportunity to do in automotive what we’ve already done in the aerospace and defense industries: we can bring more compute power into multi-core hardware environments and then run both non-critical and critical systems side-by-side, taking advantage of technologies like hypervisors or virtualization that provide time and space partitioning.
What does this mean? With proper determinism and time-and-space partitioning, it is possible to update hardware independently from the middleware and the application layer, and they can all operate on different time cycles and life cycles. That means that you can accelerate your development cycle and your deployment while reducing the cost and the complexity inside the actual hardware in the vehicle system.
This same concept can also be applied to take ECU consolidation to a whole new level. OEMs can consolidate more workloads on fewer pieces of silicon. Since the vehicle systems have been partitioned, they have the security of knowing that a breach in one compute domain will not have an impact on the adjacent domains. This translates to direct savings in cost and accelerates time to market.
In my next post, we’ll be discussing the “R” within our ACRU model.