skip to Main Content

SAP HANA on Azure

Companies that use SAP to run their businesses are facing a wealth of new choices for how they host and manage their SAP landscapes. This can include a private cloud, public cloud or a hybrid cloud mix. In the public cloud arena, it is now possible to run mission critical applications like SAP HANA on Azure. This may, or may not, be the right choice for your business. Many factors should influence your choice to place your SAP HANA in-memory database and data platform on Microsoft’s Azure public cloud infrastructure.

Defining Microsoft Azure and SAP HANA

Should you run SAP HANA on Azure? First, two quick definitions. Azure is Microsoft’s cloud platform, notable for its immense depth of capacity and breadth of service offerings. It’s a global platform. It enables you to run virtually any software stack you can imagine for a wide array of use cases. Azure naturally aligns with the Windows stack, but it is not at all limited by it.

SAP HANA, in turn, is a relational database platform that’s based on in-memory technology. With SAP HANA, data resides in the computer’s solid-state Random Access Memory (RAM). This architecture leads to dramatic improvements in database performance over traditional disk-optimized databases. The CPU can access directly for workloads like Business Intelligence (BI), ERP and other enterprise applications.

You can run SAP HANA on-premises on your own dedicated hardware, in a private cloud, at a co-location facility or in the public cloud. When you launch an SAP HANA certified Virtual Machine (VM) or bare metal server, you’re getting a system that is pre-configured specifically to run SAP HANA workloads in Azure. This means the servers will have a required amount of memory allocated to it for the purposes of operating the SAP HANA database.

There are a number of pros and cons to putting SAP HANA on Azure. On the pro side, you save on the capital investment required to stand up SAP HANA infrastructure yourself. The elastic nature of the public cloud allows customers to scale at cloud speed. Public Cloud removes some of the dependencies associated with on prem solutions and breaks the procurement cycle once and for all. Furthermore, Public Cloud allows Enterprises to utilize the latest and greatest infrastructure, ending the dreaded hardware refresh cycle. Ironically, it is the same hardware that some consider the downside. Commodity infrastructure has a stigma that suggests a lack of resiliency. Furthermore, some IT leaders aren’t comfortable losing visibility and control of the underlying Azure hypervisors.

Getting Started with SAP HANA on Azure

Running SAP HANA on Azure is not a “set it and forget it” proposition. It’s simply another deployment option for SAP HANA, and like all options it requires knowledge, training and focus. Plus, as we have seen many times, a migration to HANA, particularly S/4HANA, may involve a data migration to SAP HANA in the cloud – especially if the implementation is not greenfield. This takes work.

For example, to migrate a database to SAP HANA on Azure, you need to build a reference architecture first. Chances are, you won’t be doing a simple “lift and shift.” Depending on your use case, this may require some customizations to the preset SAP HANA VMs available on Microsoft Azure. Indeed, some of the customizations may occur to your data before you migrate it.

Microsoft makes cloud engineering services available to help you figure out your best approach. As a Microsoft partner, we have worked hand-in-glove with these teams to ensure an optimal transition for SAP landscapes.  Tasks may include things like upgrading and changing their SAP Kernel on the Azure VM.

Running SAP on Azure vs. AWS

The rich selection of application and platform options on Azure is partly the result of competition between Microsoft and Amazon over dominance in the cloud. Amazon is the primary cloud player with its massive Amazon Web Services (AWS) business. Microsoft is catching up, though. It’s a battle between two tech titans, each with very deep pockets. They’re investing billions of dollars into infrastructure and specialized service offerings—all the better for you.

We’re not going to take a side here. We support SAP HANA on both AWS and Azure. Each platform can do a superb job, assuming you approach the migration and deployment process the right way. In fact, the answer to “which is better for SAP HANA, AWS or Azure?” the answer is basically, “It’s up to you.” Some companies have an existing commitment to one platform or the other. If that’s the case, then it makes sense to follow this. If your team already knows it way around AWS, it might make sense to use that platform for SAP HANA.

Best Practices To Consider

The most fundamental best practice for SAP on Azure is to work with the right implementation partner. Azure’s self-provisioning is seductive. It looks easy, but you can find yourself in trouble right away with the best practice of optimizing your environment size for today. Right-sizing your environment means allocating the correct resource pool for your SAP HANA environment, and tight-sizing means paying for exactly what you are using. Azure makes memory, storage and compute available in big blocks only, meaning if you’re just over one threshold you could end up paying for nearly double the resources necessary. This may prevent you from getting the allocation correct, leading to sluggish performance and even instability.

SAP HANA workloads are performance-sensitive. Thus, the best practice is to collect data from relevant compute, networking and storage resources used by SAP HANA. Then, you’ll understand what your SAP HANA instance needs on Azure. This can be a challenge, though, because Azure runs commoditized hardware and only offers limited monitoring tools. It can be hard to gather sufficient insights into underlying infrastructure issues that could affect performance.

Beyond these measures, based on our experience, we recommend the following best practices:

  • Automation of all possible SAP HANA deployment workflows on Azure
  • Full-stack application monitoring, which is available through the Symmetry platform
  • Real-time custom alerts (via email or SMS)
  • Global visibility across all systems
  • Custom, automated reporting covering resource usage, configuration, performance and capacity

Limitations of SAP HANA on Azure

In addition to configuration and potential performance issues already discussed, the Azure stack has several limitations that may affect the success of an SAP HANA deployment. Azure is considered good for what is known as “any-to-any” computing. It can easily link applications and data from a variety of hosting environments, including other cloud providers and on-premises infrastructure. This is good, but it comes with a few problems.

For example, there are constraints on moving resources around. Your ability to respond to changes may be limited by the infrastructure. Nor will you automatically have visibility into your landscape. This can be remedied by use of monitoring software. Similarly, it may be challenging to meet cloud compliance goals on Azure unless you deliberately segregate sensitive information. While Azure meets all compliance standards and can offer access to their Government Cloud for enterprises that meet the criteria, little or none of this is done on an “as-is” basis.

Finding the Right Public/Private Cloud Mix for SAP HANA

In our experience, the best approach is to extend SAP HANA into Azure in a hybrid deployment, rather than place the entire solution in the public cloud. Ideally, your SAP HANA cloud solution on Azure will accommodate your current stack and your future business strategy. This hybrid cloud architecture is an area where we can help. Our managed SAP HANA cloud and expert services teams are available to work with you on configuration, administration and security in addition to hosting a hybrid cloud for SAP HANA that involves Microsoft Azure.

Contact us to learn more about how Symmetry can help deploying SAP HANA in the Microsoft Azure public cloud.

Jay Graboff, Cloud Product Manager

Jay Graboff, Cloud Product Manager

Jay Graboff is a Senior Cloud Product Manager that has been evangelizing innovation and digital transformation before there was either. With over 8 years delivering Public, Private and Proprietary Cloud, his passion and love affair with technology is rooted in what the Cloud can enable.