While writing this, I am on my way to China to meet with the team to discuss some of NXP’s i.MX products. Boarding the plane was an effort in frustration. Hundreds of folks funneling
through a single line, while other gate agents stood around and watched. I found myself wondering why can’t the airline be more efficient. Why can’t they use both ticket checkers? Why can’t they pull aside the folks with visa issues and let another worker deal with those exceptions, while the rest of us board? Why can’t they be more efficient? That is when I began to
understand why efficiency is such a key attribute to heterogeneous multicore processing (HMP).
Tasks should be separated based on processing needs. Forcing all tasks through a single core is just not efficient (just like having one gate agent handle all aspects of the boarding process is not efficient). The use of multicore processing means that the product is more efficient, less energy is wasted, which lowers power consumption and increases battery life. HMP can also increase performance by running multiple applications concurrently. Who doesn’t want to increase performance and efficiency? But the benefits don’t stop there. Allocating software to multiple cores actually increases reliability and robustness by limiting fault and/or failure propagation from software on one core to software on another. The allocation of software to multiple cores also supports failure tolerance by supporting failover from one core to another. The use of multicore processors enables architects to avoid technological obsolescence and improve maintainability. And by using multicore processors, architects can produce systems with fewer computers and processors.
The i.MX portfolio has long been touting the benefits of HMP, starting as early as 2012 with the launch of the i.MX 6SoloX. The i.MX 7Dual is another great example of an SoC built to enable embedded efficiency through heterogeneous computing. It brings many advantages including performance optimization, reduction of power consumption, and increased system reliability and security. By taking advantage of these benefits product developers can save on cost and system power while avoiding the more expensive option of an ASSP. And with last year’s announcement of the i.MX 7ULP, now is the best time to understand how HMP can improve your next design.
Ready to learn more about Heterogeneous Multicore Processing? You can download ARM’s presentation here. Or read a more in-depth view here. Or just leave your email below and we will reach out to you when we have our next HMP webinar.