Many claim the "let a thousand flowers bloom" principle and sometimes it could have an advantage. If my application is platform-dependent, significant transformation and rewriting can be meaningless. It has a professionally high risk, and due to the substantial cost, it used to fail business wise. Therefore, we may need more cloud service providers upfront. It is not surprising that after the first cloud adaptations, analysts believed that multicloud would be the right direction soon. Customers found convenient to take simultaneously several cloud service providers

We see that, even in the case of new solutions, just under the same professional and business conditions, many customers choose more cloud providers. They do so frequently that analysts may describe as a trend. Customers often appeal to their comfort. As far as the traditional IT, they do not like closure, the so-called vendor lock-in. So, for customers, the most practical solution for clouds is the multicloud. Considering an application or a solution, as I have just written, one can find professional and business arguments, but it is not valid for new developments.

However, this approach will cause severe problems in two areas sooner or later.

Data gravity

One is the well-known Data Gravity. We read articles every day that talk about PByte data volumes, billion transaction numbers. In addition to these "sizes", it is advantageous to bring data and applications that you use closer as soon as we generate incredible network traffic, which is often a considerable cost as well. From this point of view, it would usually be good if different cloud service providers would move into shared data centers, which is not realistic. Applications must be near the data.

Thus, the previously indicated multicloud trend seems to slow down, and companies and institutions often try to keep most of their applications and of course, the data they use within one service provider.

Cloud lock-in

The multicloud trend was reinforced by the fact that in many cases users are only able to use multicloud to revolve service providers actually, indicating the possibility of switching. That this may indeed be true, and the customers take advantage of this opportunity, solutions can only use the cloud at IaaS level, which would degrade them to close to VPS, mostly for higher costs. However, all the benefits that come with the platform service cannot be used by customers who have been explicitly driven to use the cloud.

Containerization is a good answer

The containerization is a perfect solution to manage the above two issues which spreads dynamically for many other reasons as well. One of the reasons why the "embedded" solutions, applications, databases ... spread in containers is the ease of portability while I use the services on a platform level. If necessary, in a particular situation I can pass it over to another cloud provider easily or forgive me, I bring it back to my on-premise environment.

With this approach, as a customer, I ensure the portability, and I can still signal the supplier the possibility of change. Due to the limitations, these thoughts were unfolded briefly as well.

Interested? You can either start or finish it with us.

It will surely go with us!