Many are wondering what the world of free software and open source will bring us for the new year, what progress will we have and what will they be the trends to come in 2019. We will surely see Linux 5.0 dispute in this new year, the technologies of containers, AI and the cloud will continue, a tireless and constant development where free software and open source projects will play a crucial role.
But there is a word that especially catches my attention and that is probably a technological trend for 2019, and of course open-source has a lot to say about it to make it possible. I am referring to the word «composability«, That is, what we know as composible infrastructure. It may seem strange to those who do not know what it is, something like the famous convergence when it emerged a few years ago and now it seems that nobody talks about it anymore, but it can be very interesting.
They are specifically those of Western Digital (although there are other companies like HP that are also very interested ...) those who are most enthusiastic about this trend, and who believe that this year will come to develop systems that allow users or customers to run compute, network and storage as resource groups from same device, provisioned as needed depending on what workloads are required for optimal performance. That is what will greatly improve current public cloud services.
This is to reduce underutilization and over-provisioning while creating a more agile data center. A very interesting idea to make a optimization and use of resources much more efficiently, according to the needs of IT and organizations. Without a doubt something very interesting that we will surely talk more about in this blog when new open source projects arise in this regard.
This clashes with the current converged infrastructure and hyperconverged infrastructure. Converged is designed for a specific application or workload, and with physically integrated compute, storage, and networking components. In the case of hyperconverged, deeper levels of abstraction and higher levels of automation are added, with software-defined elements implemented virtually, although it also combines compute, storage and network, but with a limitation of scalability.