Tibco has made several upgrades to Project Flogo, its two-year old open source software designed to provide a high level of functionality to IoT edge devices with very limited processing capabilities.
Tibco announced Project Flogo in October 2016 describing it as “a tiny open source integration engine [that] allows application and business logic to run on edge devices, simplifying IoT integration challenges, avoiding technological lock-in, and reducing costs.”
Project Flogo, Tibco claimed, “boasts one of the lightest integration engines in existence — with an average runtime footprint that is up to 20 times lighter than Node.js and 50 times lighter than Java — capable of running locally on IoT edge devices and not just in the cloud. … It simplifies IoT integration with a flow-based, web UI for building and deploying integration applications directly onto devices.”
Tibco said that, with such a diminutive footprint, it would enable device manufacturers and application developers to offload critical application compute to even the smallest of devices, “thereby extending end-user connected experience possibilities and significantly reducing the operational costs associated with architectures that require constant connectivity.”
Tibco followed this at Tibco Now in Berlin in June 2017 with the release of the Tibco IoT App Engine, a commercial, supported enterprise offering based on Project Flogo.
At a press conference in the run-up to Tibco Now in Las Vegas this week, Tibco COO Matt Quinn, said Tibco was making a number of announcements to Project Flogo to support IoT integration on devices running Flogo.
“Flogo streams allows us to do more stream-based management on devices,” he said. “One of the challenges, especially with small devices, is battery management. Polling the device takes a lot of power so you need to be able to take the events as they are happening.
“Also we want the devices to be smarter so we will be enhancing Flogo rules, which is a fully fledged rules engine sitting inside Flogo that can sit on a device. This will allow us to take action on the device either based on some sort of model, or based on constraints that have been programmed in.”
He added “We are also enhancing the work we are doing with Flogo and [open source machine learning framework] Tensorflow: running machine learning coefficients and models inside of Flogo on the devices themselves, so we are able to make very complex machine learning decisions.”
“We have also created a micro services version of JasperReports, giving us the ability to put dashboards closer to the device in a micro services based environment.
The Author attended Tibco Now as a guest of Tibco.
