Algorithms and data structures

Niklaus Wirth is a Swiss computer scientist. Among other things, he developed the programming languages Pascal and wrote the standard work "Algorithms and Data Structures".

What was revolutionary in 1976 is now taken for granted: every algorithm must be preceded by an exact data definition, and processes and data must be separated. SAP partner Data Migration Services is figuratively following Professor Wirth's path.

The ability to extract data and functions from SAP's ERP system opens up the option of decommissioning legacy systems, legally compliant archiving, continued use of all data, and an efficient and secure path towards Hana, S/4, and C/4.

Data Migration Services (DMS) has created a data structure platform for all data and processes of an SAP system to be decommissioned.

"You can make the platform an integral part of the target landscape"

explains DMS founder and owner Thomas Failer in an interview with E-3 Editor-in-Chief Peter Färbinger.

"Data and documents that are no longer needed in day-to-day business in S/4 can be stored on the platform on a regular basis. This keeps S/4 and Hana permanently lean.

Until now, new systems had to be constantly upgraded, i.e. equipped with additional resources, in order to maintain the required stability and speed. And the volume of data to be processed is increasing massively once again with the Internet of Things."

The safe path to Hana and S/4 will keep SAP's existing customers busy until 2025. The goal is clearly visible, but the path is dangerous and uncertain.

"Let's take the case that SAP prefers: going back to the standard and migrating the data via the conversion tool"

Thomas Failer tries to point out the pitfalls.

"From a data volume of five terabytes at the latest, the system shutdown would take far too long and therefore cost far too much. A weekend is then no longer sufficient. That's why my colleagues and I assume that no more than ten percent will go this route, if they can go it at all."

What do the "remaining" 90 percent do then? Thomas Failer:

"You have to solve the problem of migrating only a portion to S/4 and Hana, and be smart with the rest."

At the moment, many existing SAP customers see no other way out than to operate S/4 and the legacy systems in parallel. But that means even more effort and even more costs for operation alone.

The DMS concept also causes work, but the necessary resources are smaller by dimensions.

"While there is extra work involved in separating operational data from non-operational data and optimizing data quality"

Failer explains.

But if this achieves a data reduction of up to 75 percent, it can reduce the effort from 5000 person-days for large SAP implementations to perhaps 2000 to 3000 person-days. And the best part is, that's not a one-time benefit."

A win-win situation for the existing SAP customer and Data Migration Services:

"We are indeed growing very strongly. But the most important thing is: The move to S/4 is just creating the awareness that legacy data and systems have to be handled differently than before"

Thomas Failer defines the current situation.

"That brings us to the topic of convergence. The migration of 50,000 SAP customers worldwide - what a number! - to S/4 and Hana, but of course we can't support it alone.

That's why we're investing heavily in building and training a partner network that is now making our approach known more quickly and more broadly in the community."

Many existing SAP customers work with long-standing partners on their issues.

"That's right"

Thomas Failer also says.

"That is why, in addition to development, the expansion of our partner relationships is so important to us. This applies to resellers and implementation partners, including system integrators, as well as technology partners, for example manufacturers of ETL tools, and not least SAP itself. After all, we need to be able to dynamically map the expected changes to data structures in our tools in the future as well."

But existing SAP customers are hard to convince. Thomas Failer:

"That is indeed the case, and there is something to be said for thinking carefully before you buy. However, it is often the sheer size of the potential benefits of our approach, such as 80 percent lower operating costs, 75 percent data reduction, 50 percent less migration effort, that makes one skeptical.

Many interested parties simply cannot imagine that and how this should be possible. We have learned a lot here in the past year. We have to pick up the customers long before the actual migration, so to speak, which means very early in the planning process."

Finally, here is a use case that demonstrates the advantage of the "algorithms and data structures" concept:

"Think, for example, of machinery and equipment manufacturers."

is the example of Thomas Failer.

"There are processes that take decades in some cases. Of course, it makes no sense to keep the information that goes back further in time in the live systems.

And yet, legacy information, such as construction plans or maintenance reports, remains relevant until the facility in question is dismantled or legally mandated liability periods have expired."