Start Your SAP HANA Project on Azure And Stop Worrying about the Size, Costs, or Performance

Cloud Archivnig sap microsoft azure

Moving to the cloud as the starting point for innovation

Big Data, Internet of Things, Machine Learning and Robotic Process Automation were only a few topics that were top of mind in the recent report about Digital Transformation published by DSAG for 2019. Companies are more and more looking into these topics, however, Business Suite (ECC 6.0) is still the foundation and core in customer environment.

In order to start this transformation journey a lot of customers look to the cloud. Running an SAP system on Microsoft Azure is the perfect choice for most customers not only because of hard facts like costs, agility, performance and security, but also soft facts like the trust that a lot of customers have already placed in Microsoft.

Too small or too big – why not trying it out first? Start your HANA project on Azure with whatever size seems OK

A typical journey to the cloud begins by evaluating current SAP system parameters and optimizing it using early watch reports. Then, migrating to the cloud is like a walk in the park. But migrating to SAP HANA (be it to Suite on HANA or directly to S/4HANA) is a different story. Here customers cannot size the system as easily. In the last months I have seen a lot of HANA sizings that turned out to be completed wrong. Oversizing (and paying too much money) is a luxury you can’t afford. Undersizing and having bought hardware that is too small and requires additional unplanned investment is even more inconvenient.

Here Microsoft Azure provides an easy way out. “Go with whatever seems OK” is the fastest way to get started with a HANA project on Azure. This means that customers do not need to invest months and months in preparing the landscape. You just select one of the many SAP certified virtual machines and “give it a try”. If the system is too small or too big you just “right size” the system to get exactly what you require at exactly when you need it. For this Microsoft Azure provides already today machines with up to 4 TB memory. Very soon this size will be increased to up to 12 TB – available anytime you need it!

Benefits of offloading data to the cloud (it’s not only about the costs)

Once the system is up and running in Microsoft Azure, you can concentrate again on tuning and optimizing the operation of your SAP systems. SAP has always recommended to archive and offload data of your system. Unlike in the on-prem world where optimizing a system has no immediate effect on costs (the hardware is already bought and resources are available whether you use them or not), optimizing the operation in the cloud is visible right away on the bill.

Next to SAP native scenarios like archiving and deleting unused and non-compliant data, there are other ways to improve an SAP landscape. With tools like Datavard OutBoard DataTiering we see a great and easy way for customers to reduce the size of data in the database. OutBoard helps to reduce the size of your database through archiving and deleting data, but also offers the option to store this data in “other places”. In several projects we have seen Azure HDInsight Hadoop with Azure Data Lake Store being used for this.

Here the combination of running the SAP system in Microsoft Azure with several storage options is a perfect fit: hot data can be stored in the SAP systems database on VMs with even very-high memory requirements. At the same time warm or cold data is offloaded to other storage options available in Azure. Dead or frozen data can be deleted.

It is very important to highlight that although the warm data is not stored in the database itself, it is still available (seamlessly!) to the end-users. When users access warm data it still is there, it just does not come as fast as when it would come directly from the SAP HANA system. The technical setup is straight forward and mainly involves creating RFC destinations to Azure, configuring required authentication options in the related customizing tables and data tiering parameters in the Datavard OutBoard transaction.

With some customers we have seen system size reduction of 40% for the warm data which obviously dramatically cut the costs. Customers just had to resize the VMs to immediately benefit from the reduced systems requirements. At the same time, customers started to work with the offloaded data. This data can easily be combined with information from other data sources, like IoT. All this information then can be analyzed using analytics and machine learning services – also available from Microsoft Azure.

Putting this all together highlights how customers really benefit when SAP, Microsoft and Datavard combine their strength and tools to bring the best out of your data.


About the Author


Holger Bruchelt is a Technical Solution Professional for SAP on Azure at Microsoft, supporting customers to run SAP on Microsoft Azure. He has over 15 years of experience with SAP products.