The Internet of Things (IoT) represents a major step forward in the history of the internet – it is where the digital world meets the physical world. Every day, people use a variety of devices for a range of different things that link to an internet connection. As internet enabled devices move beyond mobile phones and laptops, an increasing amount of industries and businesses are now embedding connected devices into working activities.

Virgin Atlantic’s new fleet of ‘highly connected’ planes is expected to create about half a terabyte (TB) of data per flight each. With 21,000 Virgin flights a year, that’s 10 petabytes (PB) of data – the equivalent of 5 billion floppy disks. This is just one example of a data explosion throughout an organisation. With driverless cars, intelligent homes, smart healthcare and wearables, this isn’t set to stabilise but grow, with research house Gartner predicting that by 2020 we will see25 billion connected ‘Things’.

> See also: Why the Internet of Things is more than just a smart fridge

With clogged networks and bandwidths stretched, the world needs to become smarter to harness and embrace this web of constant connectivity to appreciate the massive data boom all these connected devices will ultimately cause.

Challenges

There are four main challenges facing businesses today:

Data dilemma in real-time

Data is the glue of the IoT thread, from the ‘thing’ to the app to the sensor, the information must be constant and of real value. If you consider that every minute of every day, Amazon makes $83,000 in online sales, email users send 204 million messages, Google receives over 4,000,000 search queries and Twitter users tweet 277,000 times, you can start to appreciate the mass amount of data being generated.

Take the connected car. With one in five vehicles having some sort of wireless network connection by 2020, a car on average produces around 20GB of data. A gigantic amount of this data is stemmed from built-in apps such as Spotify, weather programmes, Sat Nav and traffic warnings – that’s before your car is connected and serving as a WiFi hotspot.

Data intelligence at the application level is about understanding the data that needs to be transmitted and only sending what has changed. For example, if traffic is moving smoothly and there has been no change in an hour, it is unnecessary for data updates to be transmitted continuously.

Network dilemma

In the coming decade, the IoT will cause the bandwidth gap – the difference between required and available bandwidth – to balloon out of control, and unfortunately, this is growing. Businesses will start to see an enormous amount of traffic coming from a wider selection of connected devices and ill-prepared networks could be constricted further as the need to be connected ramps up.  

With this avalanche of data generated from connected devices by sending data back and forth simultaneously, we cannot expect the network to cope unless developers understand how to distribute it intelligently to reduce pressure on the network before it breaks. After all, what use is packing data into the rucksack of the internet that is already splitting at the seams?

Reliability dilemma

The internet can be unreliable and disconnect and reconnect with very little warning. Internet connection speeds can also vary between different clients and devices. The problem is that the IoT assumes the internet is reliable and able to transmit information in real-time. However, this isn’t the case. As human beings, we are notoriously impatient and this is true when it comes to our apps as we want the information we require straight away – internet connections are easily dropped and can often take a while to reconnect. The IoT doesn’t account for this.

This is particularly important when it comes to banking apps on a smartphone. Not only does the information need to be secure, the data needs to be reliable, fast and in some cases near real time and when you cannot control connectivity this is a challenge. Predominantly with banking apps, we need the correct information straight away, but the problem is, the internet is notoriously unreliable.

The issue is that apps are often developed with the assumption that the internet is always available, and that apps can casually connect and disconnect with easy all of the time, and this assumes that you  will be able to transmit all the data you need with ease.

Scale dilemma

Moving data is a huge challenge as the internet was originally designed to transport documents, not data, and definitely not real-time data. With billions of connected devices all containing sensors that move the data from one place to another, problems concerning scale could occur during peak times of the day. The rising number of connections increases the pressure on backend infrastructures and bandwidth to cope with scaling to meet demand.

The issue with scale is that organisations need to process data immediately, instead of storing and analysing it. This might be because it is unnecessary to store large quantities of stagnant information as it is not a changing state. Therefore, to solve scalability problems, developers must start to understand the data and only send changes when required. 

Strategy – how can businesses manage the data influx?

The answer lies within intelligent data distribution. The IoT requires the ability to get the right data to the right person, or machine at the right time – the internet is ultimately crucial. It requires the ability to get it right each and every time. This means sharing the right data with the right end point whatever that is.

> See also: Making the Internet of Things a business reality

The ability to understand data and only distribute tiny bits of changing state data lightens the load on the application and the network, and increases the ability to serve more users faster. If developers are able to understand data and only distribute what’s important at the application level; this is more powerful than any amount of hardware you throw at the problem. Furthermore, prioritising this data should be done at the application level where there is logic, after all, a network cannot sort data it doesn’t understand it – it just moves it.

Will it be happily ever after?

The IoT world is continuously evolving and businesses are increasingly searching for new digital opportunities to transform their business. However, the data, network, reliability and scale dilemmas are still complex issues that must be increasingly considered in a planning process.

Organisations have no choice but to re-think its approach for how they are going to distribute data across heavily congested networks to support the IoT.

If developers want to get the right data, from the ‘thing’ to the right device, machine or person at an exact time, intelligent data distribution technology must be considered to reduce the server load, and reliably support the network and all ‘things’ at scale.

Lee Cottle, director at Push Technology