Cloud Native Is The New Normal, Almost
It’s that time of year again when the IT industry is laying down predictions for the year ahead. A good many of these stories write themselves. That is, they all feature cloud computing, Artificial Intelligence (AI) and automation, mobile device centricity, IoT, and perhaps a peppering of neural networks and Machine Learning (ML).
If you’re lucky, you get a forward taster of quantum computing in most enterprise IT predictions stories too, but don’t count on that every time.
But leaving the monolithic IT vendors themselves aside for a moment, what do the implementers think? London, UK-based cloud technology consultancy and developer shop Amido says its core business comes from rebuilding old-fashioned corporate technology infrastructures in the cloud. The company itself is cloud-agnostic (it doesn’t care so much about who is running the backend … be it Google, Microsoft, Amazon Web Services or one of the contenders to the big cloud triumvirate), so its view on what’s happening on the deployment level may be of some value.
From rebuilding to “just” building
How much better would it be if Amido wasn’t spending so much time ‘rebuilding’ and could actually focus on ‘building’? To explain, if we do look at trends for 2019 and see that 2018 has been all about the now-hackneyed term digital transformation (*yawn*) to move chunks of legacy IT to cloud, then perhaps the year ahead will be freshly baked with software builds created for the cloud, of the cloud and in the cloud. We logically call this state of IT development ‘cloud-native’ programming and deployment and the industry appears to largely agree that we are on the edge of this becoming the new normal.
Amido CTO Simon Evans points to wizardry from analyst house Gartner, which estimates that 80 percent of internally-developed software is now cloud-enabled or cloud-native.
“Cloud-native applications are specifically designed to run on cloud infrastructure, hence the term ‘native’. They are growing in popularity because they deliver benefits, which include: high availability and responsiveness, plus also strong resilience and flexibility through autonomous and self-healing capabilities, such as designing for failure,” said Evans.
That’s a nice enough overview, but many of us will know that part already. So can Evans back up his proposition by substantiating it and explaining what the upshot of the cloud in residence at more enterprises will be?
Dive into a data lake
The substantiation here (and perhaps proof of real cloud-native adoption in enterprises) is found when firms operate with a data lake and apply intelligence to it. This is a term used to define a large repository of data, much of which will be held in its raw, natural or first stage format such as source system data, log files and so on.
Companies build up a data lake when they run massive web-scale cloud applications and the data lake’s incoming water channel is always open, no data is turned away. You can ‘bottle’ portions of data for consumption, analytics and business visualization once you apply the schema to it to show what its relationship is to other things.
Amido’s Evans argues that 2019 will be critical for enterprises to build a usable data lake in their organization as they continuously deploy cloud services. He insists that organizations will need to add in an intelligent set of discoverable, metadata-tagged data from all their systems, devices and services to extract value from the terabytes of structured and unstructured data (that sits in the lake) that they generate each day.
This approach to data lake information for cloud-native enterprises is argued to enable companies to run analytics, business intelligence, ML and AI and gain vital insights into new efficiencies to gain a competitive edge.
“Compared to a traditional data warehouse approach, a key principle of data lake architecture is to provide a place to land all the raw data without transformation or loss, so that any transformations on the data can be replayed at will. The challenge with this approach in an enterprise is maintaining a level of control over the landing of the data so that the volume and veracity doesn’t become overwhelming or turn into a data swamp,” said Evans.
It’s logical, it all makes sense, but it’s still a case of consultancies like Amido offering advice on what customers need to do to make cloud-native a more viable technology proposition.
The democratization of data science
Many argue that we now need to ‘democratize’ (i.e. allow all users, even non-technical ones) more of the Artificial Intelligence and Machine Learning out there to give business users the ability to process structured and unstructured data in order to help businesses make intelligent decisions and spot trends.
The industry likes to call that process the democratization of data science, but it’s basically about giving everyone usable data analytics tools, many of which will now be easier to use because we’ll be able to speak commands to them using Natural Language Understanding (NLU) … as the software itself interacts back to us with automated chatbots.
If we attempt to sum this discussion up at, we might perhaps suggest that a (positive) symptom of cloud-native is accepting the fact that your IT stack now has the potential to live in an infinitely wide and infinitely deep expanse – but you should only navigate across the expanses of the data lake created once you’ve plotted the trajectory and orbit of the ‘stars’ (we mean business objectives) above you.
Cloud-native is on the way to becoming the new normal, almost, nearly, with a few caveats … we will get there.
Self-driving cars may be closer than you realize.
This blog is provided for informational purposes only and may require additional research and substantiation by the end user. In addition, the information is provided “as is” without any warranty or condition of any kind, either express or implied. Use of this information is at the end user’s own risk. CenturyLink does not warrant that the information will meet the end user’s requirements or that the implementation or usage of this information will result in the desired outcome of the end user.