Power Limits Apps In The IoT

The industry needs to figure out how to leverage all of the data that tools generate to make sure IoT devices are exactly targeted to their task.


The applications in the IoT are seemingly limitless, but the power is one thing that can’t be.

Mary Ann White at Synopsys reminded me that a lot of the energy harvesting devices are super low power and there is a reason why they use just a simple LCD-type display. But we agree it would be so cool if we could have color LCDs that still only consume low power. Of course, I have no doubt that somebody will develop that someday.

Mary Ann noted that it’s always the user interfaces that use too much power. “It’s the reason why the Apple watch will probably drain the battery because it will have the coolest looking color display plus a user interface that’s going to be killer,” she said. “There’s a reason why it’s not on the market today probably because the battery is not that good. The more you add displays and user functionality, the more power it will take. That’s where you won’t see Apple watches being able to energy harvest but you might be able to energy harvest on a Fit Bit.”

Connected with this is thinking about how to make sure that the device is exactly optimized for its specific use case. How is that possible? Well, by leveraging the existing data available from the design tools already in use today.

This is a challenge, observed Aveek Sarkar at Ansys/Apache, especially for many large organizations because they are very distributed and dispersed. “How do you start to look at the data you are generating and gather some intelligence out of it to help guide you to faster convergence.”

“At the risk of using some cliched phrases like gate analytics and things like that, something like that will be beneficial,” he explained. “If I am putting together a 50 million instance design and running a simulation in one of our solutions, we do have access to a large amount of data in the design. We know how they are connected. We know how much power they consume. We know how they switch, how they operate, how they interact with each other. If I can leverage all of that information in a smarter way and tell you some potential issues that will arise, that’s exactly what big analytics is all about.”

“When we look at the pure EDA mindset, we’re only looking at the electronics designers but the end system requires the mechanical engineers, the industrial designers and the chip guys,” Sarkar pointed out. “All of the closed loop interactions can happen only if you share the simulation data among everybody. This goes back to the concept of faster convergence instead of the electrical engineer doing his or her own thing on the side, and saying, ‘Eureka! I’ve got my best antenna, let me go ahead and make it.’ It comes back on the mechanical engineer who says it won’t work because it’s going to be so flaky and the industrial designer says nobody is going to buy it because it looks so ugly. Versus, they all iterate in the same design environment.”

Obviously, this is a concept, but one has to think that with all of the data generated by tools, there has to be a way to leverage it better. This will be critically important to figure out to make sure IoT apps are the right ones for the task at hand.