Share
There is a lot of talk today about AI and how it will impact our society. Even though AI has been a field of study for decades, the advances in silicon today are enabling new capabilities faster than a lot of people can get their head around. Every new generation of compute devices brings a step function in what is possible. But a computer is more than the sum of its parts.
All the embedded computing suppliers today have basically the same options when it comes to silicon. It’s what we do with it that brings real value to our customers. Yes, the HPC2812 does deep learning acceleration for AI. It processes massive amounts of data to allow our customers to do great things with inference at the edge, enabling tasks like image recognition, electromagnetic spectrum analysis, and more. That’s because it features the Intel Xeon D-2700 CPU that is designed to bring this capability to rugged environments.
Where the HPC2812 differs is not in its components, but in its design. This 18-layer board puts a mind-boggling amount of componentry into a compact form factor that our customers need. One of the major obstacles in putting that much processing power into that small space is the movement of heat. These modern silicon devices use a lot of power and thus, put out a lot of heat. Removing that heat is essential to peak performance. Without key innovations in thermal management, processors can slow down at high temperatures – not something you want from a mission-critical system. You need guaranteed performance.
Another key differentiator is in the diligence we put into ensuring signal integrity. So much data moving so quickly throughout and between the 18 layers needs incredible attention to detail. The HPC2812 was recently recognized by Siemens in the Xcelerator Technology Innovation Awards when it was awarded best design overall in their annual electronic systems design competition with one of the judges commenting on the design as the “cream of the crop”. Read more about that at Siemens Electronic Systems Design & Manufacturing on LinkedIn: #pcbdesign #electricalengineering #electronics.
I’m thrilled to be a part of the embedded computing community during these exciting times. Advances in technology are moving faster than ever. Abaco has again committed to targeting an investment of approximately 10% of revenue into R&D, new product development, and product improvements next year. I can’t wait to tell you what’s next.
