Many things of the past are getting their well-deserved rehabilitation these days.
Having your own car and even driving it alone is suddenly not a high crime anymore, and even healthier than the crowded public transportation. That it is also less expensive, goes without saying.
Ample square meters of living space – sufficient heating very important – are not environmental waste now, they are good for enabling social distance and working from home.
Smartphones are again manufactured in the middle of Germany – more than ten years after the closure of the last Nokia plant, and this time with a profit.
Volkswagen hires thousands of engineers to develop 60 % of all software in house, including even an own operation system.
So, what about the brick-and-mortar basic computing, such as storage, databases, email, et cetera? Why is all that still perceived as having its ideal place in the public cloud? We already know that cloud is more expensive, and we know that it makes you run your business the way cloud providers see as right. Peter Levine was predicting the end of cloud computing to come in 2020, which is now. Why does that seem to be delayed?
There is probably only one remaining reason – decision-making safety. Same as it used to be with IBM in the past, apparently nobody has got fired – so far – for buying Amazon or other hyperscale cloud. With IBM back then, the change came through distributed client-server systems. Many small and large vendors emerged, and even IBM became one of them selling personal computers and Lotus software, but none was a safe bet anymore.
In that now famous presentation from 2017, Peter Levine cites the large volumes of sensor data and the need for local action as the main factors for computing to move away from public cloud. We should add to this list extremely inexpensive in-house hardware resources and new technologies to create and monitor bespoke applications for business. In my 2018 post I compared the storage costs, with the cloud to local difference going up from 3,5 times to 12 times in ten years. Today, the price of Amazon S3 storage is still the same it was in 2018, and the average prices for own storage went another 15 % down. APIs are becoming increasingly standardized, and low-code removes the entry barriers for businesses to create their own applications.
Maybe 2020 is still not the end of cloud computing, but there are certainly less and less reasons to not own your computing resources and applications.
At Hypersoft, we have several products to support this process. Omnicontext for technology operations monitors quality and usage of all services, cloud and in-house. It can also be used to integrate transaction data from multiple sources over long periods of time and to serve these data to other applications. Personal Files is a reliable blockchain-supported distributed storage engine for those who decide to own their data. We shall have more coming in the next months.