While networks become ubiquitous and the price of computer hardware continues to decline, more devices are being connected. Latest estimates put the number of connected devices at 13 billion, almost half of them in manufacturing, logistics, smart cities and industry applications. According to IDC, companies will introduce such initiatives within the next five years.
“The Internet of things (IoT) is already generating huge amounts of data, much of it at the edge of the network,” says Gareth Donald, solution strategist, Actifio South Africa. “The management of this data will quickly overwhelm traditional approaches within the near future. This is a critical challenge because the primary business benefit of the IoT has the potential to use the data it produces to generate business insights.”
In other words, a better method for managing data - and thus making it available to powerful analytics programs - is required. The good news is, that such technology already exists and has proven itself in the data centre.
“This approach is called copy data management, and it is based on data virtualisation. By decoupling data from its underlying physical infrastructure, the data becomes more accessible and can be better organised, similar to virtualising servers. Data virtualisation eliminates complexity and avoids data duplication at the core,” he notes.
Donald adds that traditional approaches would have segregated physical sets of data into individual application silos, resulting in massive duplication and storage requirements.
“Decoupling data from the physical infrastructure reduces the amount of data needing to be stored, making it accessible much faster. Instead of being hidden in separate application silos, virtual data is available centrally and thus to the whole business instantly.”
This reduction of complexity significantly reduces costs while increasing efficiency and ultimately speed and business value retrieved from data.
Donald says, the principle of data virtualisation is, that no matter where it was originally generated, all data is collected centrally. From there an unlimited amount of virtual data copies can be made available for any use case with instant access. “This technology is already widely used in data centres and can be applied to almost any application, platform, operating system or virtual environment, as well as private or public clouds.
“Copy Data Virtualisation is ideal to help overcome the challenges posed by data growth generated through initiatives such as the ‘Internet of Things’. As noted above, the IoT will be responsible for growing volumes of data at the edge of the network.”
Another key trend will be the use of cloud computing to host IoT data. Public clouds would appear to be the perfect vehicles for data generated by ‘The Internet of Things’ because of their scalability. “However, data centre operators and cloud service providers will need to be able to cope with the volume of physical data generated by the IoT. Bandwidth is another related issue - moving all this data around is expensive.”
The final consideration is Speed. The value of the IoT data depends on how quickly it can be accessed and analysed for insights - where expectations nowadays are towards real time.
“The strength of Copy Data Management makes it ideal for this demanding environment. Inside the data centre, virtualising the data makes it easier to manage render and reduce cost. Data virtualisation will yield the same benefits within the cloud, making virtual data copies available for immediate analytics.
“Reliable, practical solutions for dealing with large amounts of data - as required by ‘The Internet of Things’ - are important now, and their importance will increase. IDC estimates that the 13 billion devices connected to the IoT today, will have grown to 30 billion by 2020, representing a business opportunity of around $1.7 trillion. The world will truly be a connected one and the real winners will be those who have the ability to use the insights of their data to drive profit and growth,” concludes Donald.