Many Paths lead to the Cloud
The path to the cloud is a major topic for more and more German companies. The crucial question here is usually no longer "If?" but "How do we best proceed?" - The classic consultant answer to this is: "It depends."
So what does it all come down to? The procedure for each individual transition to the cloud depends on the initial situation and the specific needs of the company. This gives rise to a wide variety of transition approaches. I would like to clarify and simplify this using two analogy examples.
Let's imagine Klara: Klara has made a conscious decision not to buy her own car. She prefers to save the hefty purchase price for other investments. She was also put off by the follow-up costs for gasoline, insurance, taxes and parking space. She does not want to deal with issues such as repair, maintenance and wear and tear. Nevertheless, mobility, independence and flexibility are important to her.
Klara relies on car sharing. The monthly costs are manageable compared to buying and maintaining her own car. She only pays when she actually drives. She doesn't have to worry about maintenance issues; the car-sharing provider takes care of the service. Klara also doesn't have to decide which car she wants to drive for the next few years. She has a free choice. She decides for herself whether a compact car or a sedan currently suits her situation.
In Klara's case, the car itself plays a downstream role. The performance should be reliably consumable and using the car is the main focus.
Back to the Cloud...
The advantages of cloud applications are very similar to those of car sharing. Pay as you go, elasticity and scalability, OPEX instead of CAPEX, transparency and repeatability, self-service, to name just a few.
The higher the requirements for flexibility, scalability and volatility of an application, the more worthwhile it is to transition it to the cloud. One way of doing this is to rebuild the application on the basis of cloud native principles. Replacing the application with a software-as-a-service solution is also conceivable.
But it's not always that simple - especially when it comes to business-critical IT applications.
Frank the car tuner
Now let's take Frank. Frank is a car tuner. Over the course of several years, he has assembled his dream car piece by piece. His car is not just a vehicle. It is a valuable one-off. The value of the car lies in its history. Over the years, he has repeatedly individualized and modernized his vehicle. His car represents exactly what is important to him. The best and most suitable technologies from various manufacturers combined in one vehicle.
An off-the-shelf car is out of the question for Frank. But he is also prepared to invest time and money in the necessary modernization. Unlike Klara, Frank is very much concerned with "the" car - namely his own. His car is where the value lies.
Back to Cloud...
In many companies, we find a similar starting position. It is not uncommon for the greatest and most business-critical value of applications to lie in the specification and individualization that has historically been carried out.
These highly specific individual applications can be found in every IT landscape. Not all of these applications can be moved to the cloud without further ado. Older applications in particular often require costly and time-consuming conversion before they can run satisfactorily on a cloud platform. The offerings of cloud hyperscalers, on the other hand, rely on standards - there is rarely any room for negotiation in operating and service models. However, many companies need a customizable environment for their special and individual solutions and the associated services.
Replacement or substitution with standards is out of the question, costs the competitive advantage and or is considered extremely risky.
In this case, the existing (partial) landscape can be raised 1:1 to a modern cloud level. It is possible to connect an existing data center with the data center of the cloud provider via a virtual connect, also known as a layer 2 stretch. Applications can then be transferred individually. IP addresses do not have to be changed in the process. Smooth continued operation is therefore not jeopardized. In short, this makes the successive transition to cloud technologies possible.
"So it depends."
Klara and Frank are just two examples. The starting situations that prevail in companies are as diverse as the companies themselves. Transition consulting is therefore the be-all and end-all of any move into the cloud. Only those who know their starting position can make efficient cloud decisions.