You may have noticed lately that Internet of Things (IoT) noise has lowered a notch as the sector starts to get its teeth into making viable solutions that will be a really strong foundation for the revolution to come. Just this week a survey by security folk Webroot and data centre specialists IO suggests that 54% of UK businesses are looking to hire a Chief IoT Officer (CIoTO) within the next 12 months to help navigate their entry into this disruptive new market. ComputerWeek reports that a further 68% of UK business leaders are expecting their IoT investments to garner profit this year, and a total of 94% of businesses are making preparations for the IoT. With lofty figures like these flying around, it’s little wonder that developers are keen to prepare themselves for the interconnected revolution.
The latest news in this push to connect all the things is this week’s open sourcing of IBM data analytics tool Quarks. As we’ve seen with the Big Data phenomenon, being able to extract and stream data is one thing – actually spinning these torrents of information into anything of value is considerably more difficult. With IoT data, data must be streamed over communications networks before it can be analyzed by a centralised application, ramping up costs and adding a considerable time lag. Operating as both a programming model and microkernel style runtime, Quarks – which can be embedded in gateways and devices to deploy timely edge analytics on data streams and devices like the Raspberry Pi – is intended to address this hole in the market.
Quarks comes packed with Java APIs for performing analytics via a per-event streaming paradigm, and connectors for MQTT, HTTP, JDBC, File, Apache Kafka and of course the Jeopardy winning IBM Watson platform. Its development mode includes a console with a viewing graph for running applications, as well as a testing mechanism for its applications which is compatible with assertion based testing systems like JUnit.
IBM hope that by opening up Quarks to the wider developer community to tinker with, the tool’s ability to perform local analytics on the huge variety of edge devices on the market will come on in leaps and bounds.