“Cloud first” vs. “All eggs in one basket”

Published on August 5, 2017, on LinkedIn

Enthusiasts of ‘digital revolution’ run between two slogans – “Cloud first” vs. “Digital first”. I personally think that both slogans are rather a buzz because in the market economy the first was and is the business.

I’d agree with “Digital first” because it is a general statement about implementation of business solutions, i.e. it clearly defines the approach without prescription on how to do this leaving a room for changes and innovations.

A movement “Cloud first” enforces us to accept and go to a bottleneck constituted by Internet communication channel, which is outside of Cloud realm. Cloud concept assumes that Internet is always up and running, but it is just another technology with no one responsible for its quality and serviceability. This means that each company using Cloud also gains an additional risk that it had not before.

Let’s assume you are a public organisation in the UK and you move your IT in Amazon Cloud. Your benefits are obvious – you can now save on IT infrastructure, software and hardware, as well as on its maintenance. You are happy now. Well, this is a bit late but the question business continuity when dealing with Cloud still remains not answered.

This is not about Cloud elasticity, this is about Internet connectivity, that according to Amazon, its Cloud is not responsible for. For example, for the last year, Amazon went down from the consumer perspectives three or four times, globally or locally or for particular fundamental service. The business continuity question, therefore, is: may a public/government organisation afford being paralysed for a day or two because it technology resource suddenly appear inaccessible?

Using different Cloud providers does not eliminate aforementioned risk – they likely use the same Internet connectivity network even if they a physically situate in different territories. Really reliable solution here is to have a redundant Internet connectivity networks (which is not up to Government to manage) or have a resilient means for the business activities, including related automation and data, that can be quickly engaged in the case of failure in the connectivity to Cloud.

Yes, these resilient means significantly decline efficiency of the Cloud use for now. That is, before going to Cloud based on the “competition mentality” it is necessary to calculate the benefits and losses when the risk materialise. For example, a few years ago, there was a very heavy snowing in London. The snow was left in the streets for more than week because, as officials explained, the city government did not anticipate such snow and did not allocate related funds to hire snow-machines to clean the streets. The city was stuck. The losses for just a week exceeded needed funds many times. The same is with Cloud.

My opinion is that until we cannot guarantee a reliable access to the public/Government systems/data in Cloud, transition into Cloud has to slow down and taken with all necessary precautions. This has to be the Government strategy in “digital first” regarding Cloud adoption.