The cloud services market has had a profound impact on the data centre industry. Many of these facilities now use virtualisation and automation technologies that, in conjunction, deliver efficient infrastructures.
Configuring solutions like VMware and OpenStack enables IT service providers to offer a wide variety of web-based provisions, from scalable storage to application development platforms. The data centres behind these services are built to support millions of transactions, all of which are initiated by remote machines.
What happens when the number of devices accessing cloud resources expands dramatically? What if the data those machines provide is complex and sizeable? These questions are associated with the Internet of Things (IoT), or the concept of billions of devices connecting to the internet and communicating with one another.
Data centre resources and the IoT
According to a 2014 report from Gartner, approximately 26 billion "things" will make up the IoT by 2020. Many, if not all, of these devices will connect with centralised management systems capable of processing their data and delivering services associated with them.
For example, a network of industrial measurement instruments at an aluminium manufacturing plant will collectively send data to a cloud-based application. That system will then analyse readings to inform managers of equipment damage, production rates and other factors.
However, these capabilities aren't going to come without growing pains, particularly in the data centre. Gartner Vice President and Distinguished Analyst Joe Skorupa outlined the challenges the IoT will present to data centre managers.
"The enormous number of devices, coupled with the sheer volume, velocity and structure of IoT data, creates challenges, particularly in the areas of security, data, storage management services and the data centre network," said Mr Skorupa.
Managing data efficiently
Like many IT concerns, information is presenting the majority of the challenges associated with the IoT. As Mr Skorupa mentioned, smart devices produce an incredible amount of myriad information quickly, obligating professionals to exercise certain server management practices. However, achieving effective data management may involve moving past the data centre.
Cisco recently released an interesting white paper titled Fog Computing and the Internet of Things: Extend the Cloud to Where the Things Are. In the piece, Cisco maintained that the conventional cloud computing approach, which involves moving all information to the data centre, cannot meet the requirements of a vast network of multiple devices.
What are those stipulations? According to Cisco, in order for companies to handle data efficiently, an IoT-based ecosystem must do the following:
Addressing security and resiliency
Although the topic may seem tiresome to IT professionals, they shouldn't ignore the security implications of the IoT. Protecting information not only involves applying encryption algorithms, but also entails ensuring devices are available.
For example, businesses will need to develop more responsive disaster recovery strategies. If devices can't access data centre resources because of unscheduled downtime, a web-dependent operation may suffer as a result.
Ovidiu Vermesan and Peter Friess' "Internet of Things – Converging Technologies for Smart Environments and Integrated Ecosystems", noted that IoT environments will require extensive access control and monitoring applications. While authorisation programs will dictate how can conduct administrative tasks, surveillance tools will scrutinise situations to detect threats. The latter applications will likely be hosted in a data centre, operating over virtualised servers and gathering information continuously.
Distributed computing may transition some processing responsibilities away from large data centres. It's also possible that companies managing a large number of devices may set up server cabinets that collectively run cloud infrastructures. Supporting such systems may prove challenging, provided they lack the personnel capable of managing them.