Over the past three decades, the way and the amount of data that people use has increased exponentially. The Enabling NFV-based Service Cache in Edge Clouds project is looking at ways to improve the ability to transfer substantial document data and reduce the strains on existing networks.Dr Shan-Hsiang Shen, from the National Taiwan University of Science and Technology, notes that market commentators predict that cloud-computing traffic is likely to rise ‘nearly fourfold by 2020, increasing by 3.9 zettabytes (Zb) per year to total around 14.1Zb per year by 2020 and on from there’. The fifth generation (5G) network system is already well into its planning stages and with capability studies on networks are underway. This next generation of data transport hopes to enable not just megabit per second (Mb/s) but to extend into gigabit per second (Gb/s) speeds too.Shen says that edge cloud computing is becoming increasingly important to the delivery of data as it shortens the time for retrieval by keeping the data on relatively local servers located in the edge cloud rather than a user device requesting it from a different computer or server system – which may in reality be on the other side of the world – and transporting the data from there. ‘In addition, the potential network required for retrieval will most likely not be linear, putting strain on the system and lengthening retrieval times as the data may have to travel to different server locations on its journey.’ However, while cloud computing offers some respite, the sheer mass of data is growing to enormous levels and even new solutions in this space are required.Shen understands the problems of data flow, particularly when applied to highly complex networks. He points to the situation where a user may want to watch a video available on CNN or YouTube. By requesting it via a webserver a flood of requests are made until the original request reaches the content location and starts the transfer. The server then relays the content back to the requesting location while caching it along the way. If a second request location also demands the same content but rather than requesting it from the content location it may be taken from the data that has already been cached in the network. This is known as Content-Centric Networking (CCN) and is a huge step forward from the previous IP-based, host-oriented, architecture, but are still relatively slow. Shen is suggesting that much more emphasis can be placed on cloud computing, and its newest iteration, edge cloud computing, which he terms as a service centric networking (SCN) and explains is a more advanced version of CCN: ‘Compared with CCN, SCN can support more applications. CCN can only cache data’’ closer to end users, so it cannot support the applications with more interactivity locally’.