Build Microsoft has wheeled out open-source software that wrangles Internet of Things devices and beams data to and from Azure.
It's a stack of code that runs on hubs and gateways that sit between small IoT gizmos and backend services hosted in Redmond's cloud. It basically lures engineers and techies into Azure.
Announced during the technogiant’s annual Build conference, held in Seattle this week, the MIT-licensed Azure IoT Edge code tries to run all of today’s latest marketing buzzwords – artificial intelligence, edge analytics, etc – at the network edge, bringing compute a few theoretical steps closer to your endpoints.
It's a follow-on from last year's Azure IoT Gateway software libraries, a development kit for interfacing devices to services running on Microsoft's public cloud. To get the idea of Azure IoT Edge, see Redmond's diagram below – the box in the middle is where the software stack runs, on some sort of hub, relaying data to and from the cloud and lightweight gizmos, doing any necessary processing before shipping information in and out.
Yes, it run on Windows or Linux. We touched on the technology earlier, here, in our Build conference roundup.
“It makes sense to move some of the cloud intelligence out to IoT devices themselves,” said Microsoft's Sam George on Wednesday.
According to the Windows giant, the software is supposed to help manage and control devices, adding that it “greatly simplifies the complexity of deployment, configuration, updates, and health monitoring of IoT solutions involving a varied set of devices.”
It also keeps on working when there is “intermittent” cloud connectivity by re-sync’ing when connectivity comes back up.
As mentioned, you have a choice of Microsoft's operating system or open-source Linux, and it'll run “on devices even smaller than a Raspberry Pi with as little as 128MB of memory,” we're told.
It's too hefty to run on a bog-standard IoT sensor or controller, which tend to use small ARM-compatible cores, a real-time OS and limited memory, power and performance. Instead, the code is aimed at internet gateways – perhaps next year's home broadband hubs – that sit between these smaller gadgets and the wider world, and do some intermediate processing on behalf of their endpoints. Exactly what that hardware is, well, it's up to you to figure that out.
Edge computing is the latest thing the big boys have jumped on while trying to take hold of the IoT market. To pull a random example out of the air, HPE’s Edgeline servers pack 16-core Intel Xeon CPUs plus up to 4TB of storage into a footprint about the size of the average desktop – quite a far cry from a stock Raspberry Pi. No doubt owners of USB toothbrushes will be jumping for joy at this leap in innovation. ®