As the number of internet connected devices continue to explode, organisations need to understand the convergence of IoT and data analytics.
The number of IoT devices is expanding at a significant rate. All these devices are capturing and relaying different data sets, which are giving both private and public organisations insights that would have previously been unknown.
Utilising all this IoT data, taking advantage of it, requires an effective analytics strategy that responds in real-time.
Taking advantage of IoT and data analytics
“One way that organisations can check whether they are capable of moving into IoT and data analytics is to look for third-party data which may enhance the data they already have.
“Once sourced, they then need to see whether they can easily combine this data with the originally stored data and put this into an analytics platform to test the capability.”
After finding the right partner, organisations can begin to see whether there is an actual business benefit coming from it.
In order to do this, Ruffley explains that organisation’s don’t need to invest heavily. “In fact, this can be done on generic platforms such as Google, where the original investment is very low. Just by doing a simple practical exercise like this, companies can see how ready they are to take advantage of IoT and data analytics, and whether there is business benefit in doing so,” he says.
Managing data: 5G and data fabrics
He points to 5G’s roll out as being an enabler for an “unbridled level of data flow and a huge uptick in new edge computing capabilities. This means that the challenge of organising data and making it valuable has never been greater.
Using an example of a well-known car company, Watts explains: “[they] built up about 14 petabytes of storage over several years, but in just the space of a few months, its newly deployed driverless car program created an additional 4.5 petabytes. It’s the breadth of potential deployment mediums, interoperability and data formatting hurdles that makes analytics a real headache. The task of seamlessly connecting the edge to core to cloud in an actionable and accessible way is paramount due to the nature of distributed compute.”
He identifies data fabrics as a potential solution — this “enables a single, standardised data management and storage solution that works across different architectures and platforms, a well thought out data fabric strategy facilitates the link between the data platform, data lakes, APIs and key analytics products that will be used at a higher level to generate meaningful IoT insights and actions.”
“IoT and data analytics are complementary technologies” — Pilgrim Beart, co-founder and CEO of DevicePilot
IoT data at the edge
Suman Kumar, director of digital transformation at CGI, reiterates that “there is an emerging pattern of IoT data processed at the edge before getting stored centrally in the cloud where big data and AI technologies further transform these data sets into actionable insights.”
He continued: “Some example use cases are predictive maintenance alerts in manufacturing, utilities coupled with machine learning algorithms to automate the scheduling of maintenance tasks. The insurance industry is looking to gather more and more real-time data (e.g. telematics, smart buildings) to develop a better understanding of risk using AI. The logistics industry is implementing initiatives to get a real-time view of products in the supply chain. Local authorities and Central governments also want to develop smart cities to improve traffic management, parking and other citizen services.”
Driving value from IoT data
Like Watts, Nelson Petracek, global CTO at TIBCO Software, understands that the convergence of IoT and data analytics will be a critical aspect of how organisations can take advantage of IoT-driven computing and computing at the edge.
He says: “Digital transformation includes the ability to make relevant decisions in a timely and contextual fashion, and the combination of data generated by IoT devices and analytics can help provide this capability.”
However, simply collecting IoT data is not enough — “Organisations need to turn this data into value in both a batch (using traditional analytics) and real-time context. It is also not desirable, nor possible in some cases, to do all of your processing at the enterprise level (in the cloud or data centre, for example).”
As is the nature of IoT devices, decisions will often need to be made in a localised fashion, including on the device itself, and these decisions will be largely driven by models derived from analytical processes and historical data.
“The ability to make the edge ‘smarter’, offload compute workloads to the edge for more efficient processing, support localised or independent/disconnected processing, reduce decision latency, and reduce data transfer requirements are all benefits that may be applied to almost any vertical,” continues Petracek.
“Analytics, and the operationalisation of analytical models and pipelines, presents a huge opportunity to organisations, especially given the level of real-time information and context that IoT can provide.”
Identifying the challenge of IoT and data analytics, Bernd Gross, CTO at Software AG, said: “There are still businesses with an old school mindset when it comes to data — they see it as something you look at once a week in a meeting; you see interesting patterns and that’s that. Not only do all businesses need to be equipped to act upon data as soon as possible, the ticking clock is going to get louder and louder. IoT data is perishable, its value degrades over time. We need to see it clearly and quickly to be able act on it decisively.
“The challenge is that this data rarely sits in one place. It could be that the infrastructure is historically siloed, or you operate in a multi-cloud environment. Either way, businesses need to integrate their data. It’s the only way analytics can see ALL the data it needs, and ultimately, any data is useless without analytics.”
Look to multiple vendors
But, according to Beart, “most analytics tools don’t natively understand IoT telemetry, although there are certain analytics tools developed specifically for IoT, such as Service Monitoring tools, which help to keep a device estate working well and ensures that it delivers value to the customer.”
He uses the example of an estate of vending machines, where analytics can take telemetry data and turn it into a list of goods that need replenishing and also schedule service technician visits.
However, Beart goes onto explain that “the days of all-in-one IoT platforms which include analytics seem to be behind us. No one vendor is good at everything, so integrators today mix-and-match services and solutions from multiple vendors. Broadly each IoT vendor specialises in devices, comms or cloud and then additional services can be plugged in, including analytics.”
Author: Nick Ismail