With estimates as hyped up as 50 billion devices expected to be connected by 2020, the internet of things (IoT) will be the most significant change to the way data is managed in a generation.
And that has enterprise data centre operators going back to the drawing board.
The surge in the range of devices and sensors that create and send data means the IoT revolution will bring huge data inflows to centers that will need to be stored, analyzed and used. In fact, the very direction of the data flow represents a significant shift from the past, with centers traditionally feeding data-hungry networks.
But the big promises and opportunities of the IoT means that data storage costs are expected to balloon with huge amounts of inbound traffic.
Real-time (and order) is of the essence
While the discussion around the IoT is centred on the possibilities for healthcare, education, public transport and other improvements in civic life, IT managers will need to consider the time- and order-sensitive nature of IoT data.
Whether it’s sorting through peak-hour public transport or remotely monitoring elderly citizens’ health, getting accurate and reliable data on time and in order is important so that automated decision-making has all the available information.
That means data should be sent in real-time, or time stamped in batches.
Getting the architecture right will be key to successfully hosting the rapid connectivity that is a core feature of the IoT. To avoid storage and networks creaking under the weight of the data, and to account for yet-to-be-conceived IoT applications, data center operators will have to completely rethink systems and management.
One consideration might be the move away from big data centers to smaller pockets of storage across a range of sites. Coping with the enormous number of devices and sensors will be a challenge that will almost force a return to a more distributed network. But that might also mean a better handle on security.
A host of other planning considerations will need to come into play, including network bandwidth to cope with increasing frequency of the packets of small data, how that data will be backed up and the increased processing required to sort through all the data that comes into the network.
While the IoT promises to drive a new industrial revolution, the gauntlet has been thrown down for enterprise data center operators to meet the challenges of its introduction.