Week 4: Data & Analytics – From On-premises to Cloud
Summary: Many companies want to move their data and analytics solutions to the cloud but are still on-premises to a large degree. So, how to get from current to future architecture? In this article of the series “Give Data Purpose Weekly” we talk about architecture, a bridge, spaces and an invitation to the Data Science teams . From Kay Patzwald and Lars Jakob.
What does an Upper Lusatian House have to do with data & analytics?
Some time ago while we were debriefing a customer workshop in Blauensteiner in Vienna, we came up with an analogy about transformation from old to new.
Kay has helped a friend to renovate his newly purchased Upper Lusatian house (Umgebindehaus in German). A special type of construction, which is common in the region from Upper Lusatia to the Elbe Sandstone Mountains. Timber framing is combined with a solid rubble stone construction. A structure, the “Umgebinde”, a wooden support system, which runs around the house, is the basis for the upper levels. The house is 200 years old. You can live in it. But the desire to modernize is apparent. As other houses in the region show, there are wonderful combinations of established structures and adaptions with modern technologies and materials. Esthetically an absolute eye-catcher. Functionally top notch. In such a house, you would like to live.
It didn’t take long until we drew parallels to our customers challenges to modernize data & analytics with SAP Business Technology Platform.
A modern data architecture
Modern data architectures at present are combining classical enterprise reporting and advanced data analytics. The solid structure, the fundamental KPIs, are changing slowly. Meanwhile, requirements for data availability and insight are growing rapidly. And there is a need to support the integration of all data and bring everything together. Here the cloud comes into play.
Cloudify your Data Warehouse
In comparison to cloud solutions traditional on-premises Data Warehouses are less agile. Whenever structural changes in the data model are necessary, solution specialists are required for the implementation. Additional data sources can’t be added without support from the IT department. Increased storage requirements may lead to additional hardware. Waiting times of months for some requests are not an exception.
Investment protection and BW bridge
On the plus side these requirements are proof that the data and the models in the Data Warehouse are valuable and in need for the business. Especially in the cloud. Yet, a completely new development of a Data Warehouse is often hard to accomplish during ongoing operation. Nevertheless, there are customers who tackle a greenfield implementation. And SAP is the right partner to do that. SAP BW customers, who are looking for redesign instead of a completely new development, can make use of BW bridge in SAP Data Warehouse Cloud to master the challenges. BW bridge accelerates the move into the public cloud by allowing to reuse existing investments and skills. Just as you would imagine. A bridge that brings you from on-premises to the cloud.
SAP Data Warehouse Cloud adds modern features. Cloud potentials like scalability are combined with new ways of data modelling and collaboration.
One example are the possibilities of the integrated and simple graphical modeling interface. Less technical users can use the no-code environment to model, visualize, and share data. And it also includes the integration of additional data sources. However, the connections can be established by IT and shared with Business. And thanks to more than 160 predefined connections for SAP and non-SAP solutions the effort is minimal.
All that without losing the business semantics and the meta data of the KPIs. A solid foundation for analytics for example with SAP Analytics Cloud.
In SAP Data Warehouse Cloud users are working in so-called spaces. You can imagine a room in a house you’re only allowed to enter with your access card. In the space are your tools and your data. If you need additional data, you can get them on your own or the room service will provide them for you.
The spaces allow users to work in their secure environment according to their role, but also to share data with other users or departments.
SAP and Data Science
The modern Data Warehouse is also occupied by other inhabitants. Many companies have established Data Science Teams. The goal: gain additional insights from data and build data-driven applications.
And the most relevant data for steering the company is found in the ERP system, the CRM system, and the Data Warehouse.
The space concept of SAP Data Warehouse Cloud also meets the requirements of the Data Scientists. Here you can find data with business semantics. The effort to copy the data table by table and then add semantic again can be omitted.
Because of the openness of the SAP solutions to work with non-SAP tools on data, the work can start directly. Familiar tools like Jupiter Notebooks can be used. And results like for example python algorithms can easily be integrated in existing business processes. This task is performed by SAP Data Intelligence, another SAP cloud solution, which integrates very closely with SAP Data Warehouse Cloud, SAP BW/4HANA and many other SAP solutions.
And this is where the wheel comes full circle. From analyzing enterprise data, over pattern recognition in granular raw data and building algorithms, to bringing everything to productive use.
Critical success factors – then as now.
And following our example from the beginning. 3 elements. A solid basis, modern features, and a supporting structure to hold everything together are critical for success.
Find out more in our next blog posts.
Do you also want to give more purpose to your data?
Join us for our #GiveDataPurpose Intelligent Data Workshop where we work together with you to align your business goals and your data management capabilities in order to provision the right platform.
Have a look in our popular paper “12 ways to give data purpose in a multi-cloud world”, where we provided a quick preview into 12 main ways to overcome typical hurdles that keep companies from using their data.