By Nick Flaherty www.flaherty.co.uk
The recent article on edge computing by US venture capital fund manager Peter Levine made for interesting reading, and a wry smile.
Edge computing, like Cisco's idea of 'fog' computing, means putting more processing power at the edge of the cloud, which helps to reduce the power consumption of the data centre. This is a message that Chris Rowen, CEO of Tensilica and now a key strategist at Cadence Design Systems, has been highlighting for a while now, and brings more performance (and power requirements) into the embedded market.
This is great news for embedded designers. While the focus for investment in recent years has been apps and software containers on data centre hardware, the move to edge computing will need a much broader base of software running on higher performance embedded systems. That hardware is becoming available with quad and octo-core processors based around the ARM and MIPS architectures - the investment will be in the software, from machine vision to artificial intelligence.
This will drive some key changes that we are already starting to see. Google's move to combine its Brillo operating system with the Android infrastructure for software development and app discovery and the Weave cloud links is part of this move. The challenge is providing the right level of performance - including more complex databases - in embedded devices. The first step will be more computer vision, requiring more dedicated processing blocks in the embedded processors, and much more focus on the skills of the embedded designer.
No comments:
Post a Comment