Running webservices on resource-limited embedded platforms
Thanks to IoT, the use of connected embedded platforms is exploding. How do you run webservices from slow CPU with few megs of RAM?
Smart home, IoT, Arduino,... all these fancy buzzwords have one thing in common: low-cost hardware gets connected to the Internet. To integrate these connected devices with an existing system, some method of communication needs to be developed. Using HTTP REST API seems easiest and most straightforward way to do it. However, you cannot really run LAMP or node.js on a device with 120 MHz CPU and limited memory.
Experience from the Real World
The talk will cover struggles we had in Geolux when we first tried to run a simple webserver on an embedded platform used in our radar sensors. We will present the first framework we have selected, why it did not perform well, and how we managed to quickly switch frameworks/languages, while preserving most of the backend code.
We will try to debunk one commonly held myth related to programming languages and web development! Warning: heated discussion expected!
Saturday, 2017-10-07 @ 17:25
> Skill level: intermediate
> Duration: 25 min
- MSc in computer engineering, Faculty of Electrical Engineering and Computing, Zagreb (2004)
- Started with PhD research, published few scientific papers, put PhD studies on hold until day expands to 30 hours
- Teaching assistant at Faculty of Electrical Engineering and Computing, Zagreb (2004-2006)
- Junior software developer at SkyMobileMedia, implementing various communication protocols (2006-2007)
- Software developer at RIZ Transmitters, writing digital signal processing algorithms for high-power transmitters (2007-2009)
- Senior Developer at Mireo, worked on GPS turn-by-turn satnav software (2009-2016)
- Co-founder, CEO, lead software developer, ninja/samurai/evangelist/warlock/wizard/genius at Geolux (2016-) (http://www.geolux-radars.com)
- Sensors, data acquisition systems, digital signal processing
- Embedded systems (low-level hardware, drivers)
- Software optimizations for resource-limited platforms