How to take advantage of the Internet of Things

With so many pundits, commentators and experts rushing in to say "Hey guys the Internet of Things is going to be a really big thing" I thought it'd be fitting if I had a quick chat with Sean McAvan, managing director of NaviSite Europe, all about the Internet of Things (IoT), how to take advantage of IoT, pitfalls to avoid when building an IoT platform and why the cloud and IoT are joined at the hip.

With the Internet of Things tipped to be the buzzword of next year, what steps should businesses take to make advantage of the Internet of Things?

Enabling internet connectivity for devices introduces a great deal of functionality and convenience for their users, and potentially delivers huge efficiencies from centralised data analysis and management. But these advantages also create problems for those who design and manage the supporting systems. Fortunately, the cloud offers an optimal platform for the back end.

Because of the limited space for computing or data storage, these end devices tend to be simple data gatherers or displays for data manipulated at another location. All of that data storage and processing can occur within a cloud environment where developers have free rein to build out according to their needs, and system administrators can apply all of the necessary controls to keep it running smoothly and to keep the bad guys out. Cloud provides these basic tools without users having to worry about capex or data centre space.

But what about the sheer number of end devices? As the number of "simple data gatherers" exponentially increases how are businesses going to cope with the scale?

The growth in connected devices is phenomenal. This type of growth is usually thought of as a good problem to have, but can be a significant issue for those who host the core applications within their own data centre. Cloud platforms allow for easy scaling, both horizontally (adding more servers) and vertically (adding more resources to existing servers). When coupled with the "pay for what you use” model in most cloud platforms, this speed of deployment means that an organization no longer has to make significant investments in anticipation of success, but can rather keep their infrastructure just ahead of the demand curve and minimize financial outlays.

Many of these devices have long lifecycles once purchased (how often do you buy a new bathroom scale?), but consumers have become accustomed to frequent updates for online applications. That Internet-connected scale may initially just send your weight to a web portal for simple viewing, but what if the scale vendor could also send it to your doctor? What if they had an application that allowed you to have weight-loss competitions with your friends? Innovation won't occur as quickly on the actual scale as it will at the hub site.

Developing these new applications requires the ability for developers to have quick access to copies of production environments, to build and destroy testing and quality assurance (QA) environments, and then to push that new code to production. Cloud provides a flexible and agile backbone on which all aspects of an application can run.

Lastly with so many businesses looking to implement a private/ hybrid cloud infrastructure what are the main pitfalls that businesses should look out for when implementing these systems?

Plan to optimise your hybrid cloud environment, start small and scale. It’s important to audit the systems that you have and to optimise your hybrid cloud configuration so that the right resource is matched with each workload. A hybrid approach gives you the option of scaling resources for each workload and running applications on whichever platform is best suited for that workload: a highly dynamic app with unknown spikes may be best supported in the public cloud while a performance-intensive application may be better off in a private cloud. Data can be located where regulatory or security requirements dictate. Gartner recommends starting a hybrid project with a small pilot, getting comfortable with the ins and outs of the hybrid model, then rolling it out further across the organization.

1. Keep scalability in mind right from the start: while the pilot project may be small in scope, the infrastructure deployed should be ready for growth and capable of delivering an ROI (return on investment) within a defined time frame.

2. Look at Your Network: A hybrid strategy requires a close look at your enterprise network for bandwidth and scalability. With a hybrid strategy, companies will be relying on their network to ship large amounts of data back and forth (as opposed to more episodic dumps and updates), putting far more demand on the network than previously.

3. Culture shift: Ultimately, some of the biggest challenges in moving to a hybrid infrastructure are less about the technology and more about management. Most IT departments have a culture centred around control and technical expertise and now has to accommodate a more collaborative, service-oriented approach for the provision of automated, self-service IT capabilities via the cloud.

Huge thanks to Sean for chatting with us, you can follow him on Twitter @SeanMcavan.