The buzzword of the moment!
The term ‘hyperconvergence’ usually represents tools that can generate Cloud infrastructures in an almost totally automatic way, demanding just some starting operator’s parameters. I say ‘usually’ because this term is still extremely young (we can find the first traces in 2012): still today it is used in different ways based on the field it is related to, like storage or network scalability, provisioning systems of virtual desktops or datacenters managing devices.
Actually, ‘hyperconvergence’ can refer also to some architectures or virtual or physical appliances. The fundamental difference with respect to a classical system is that a hyperconvergence architecture is conceived to be basically modular, with all the pieces created to work together and to be managed by a unique entity, proving to be extremely convenient and practical above all in the management of the scalability. Pieces are often physical devices that can autonomously manage networking, storage and the virtual devices provisioning. In case of need it is necessary to tie a new piece to the others and it can be immediately used. In classical systems, pieces are conceived to collaborate all together but, given the heterogeneity of the market, it is always necessary a configuration/mediation from the operators to have all the pieces working at the best level.
The first foundations of hyperconvergence architectures can be found in the storage systems: nowadays these systems can be scaled in an extremely easy way, just by connecting all the appliances. If the pieces use the same technologies (that is: they are from the same manufacturer) they will configure in the best way almost completely by themselves.