SAP on Hyperscalers: Making the Right Infrastructure Decision

The question of where to run SAP — on-premise, in SAP's own managed cloud, or on a hyperscaler — has become one of the most consequential infrastructure decisions an enterprise can make. And within the hyperscaler category, the choice between AWS, Microsoft Azure, and Google Cloud carries implications that extend well beyond server specifications and pricing calculators.

This article offers an independent framework for thinking through the decision.

A decade ago, the hyperscaler decision for SAP was primarily a technical one: which platform offered the right instance types, the right storage performance, and the right network architecture for HANA workloads.

That calculus has changed. Today, the hyperscaler decision intersects with:

• SAP's own commercial relationships, which influence what is available under RISE with SAP and at what price
• Enterprise-wide cloud strategy, as most large organizations have existing commitments and discount structures with one or more hyperscalers
• AI and data platform strategy, as each hyperscaler offers distinct AI services that may be relevant to the broader SAP data and analytics roadmap
• Regulatory and data residency requirements, which vary by hyperscaler and by region

AWS has the longest track record with SAP workloads and the broadest global infrastructure footprint. AWS and SAP have a deep partnership, and AWS offers a comprehensive set of certified instance types for SAP HANA, including purpose-built instances optimized for in-memory workloads.

For organizations that are already significant AWS customers, running SAP on AWS offers the opportunity to consolidate commercial relationships and leverage existing enterprise discount agreements. AWS's breadth of services — particularly in data, analytics, and AI — also makes it a strong platform for organizations that want to build SAP-adjacent capabilities on the same infrastructure.

The consideration: AWS's SAP practice, while mature, is largely delivered through partners. Organizations should evaluate the quality of the specific AWS partner ecosystem in their region and industry.

Azure's primary advantage for SAP customers is the Microsoft ecosystem. For organizations running significant Microsoft workloads — Office 365, Teams, Dynamics, Power Platform — Azure offers native integration that the other hyperscalers cannot match. The SAP and Microsoft partnership has deepened significantly, with joint development on integration between SAP and Microsoft's data and AI platforms.

Azure is also the natural choice for organizations that have made significant investments in Microsoft's enterprise agreements, where SAP workloads can often be accommodated within existing commercial structures.

The consideration: Azure's global infrastructure footprint, while extensive, is not as broad as AWS in certain regions. Organizations with significant operations in markets where Azure's presence is thinner should evaluate this carefully.

Google Cloud's differentiation for SAP customers is its data and AI platform. BigQuery, Vertex AI, and Google's broader data infrastructure offer capabilities that are genuinely differentiated for organizations that want to build sophisticated analytics and AI applications on top of SAP data.

Google Cloud has invested significantly in its SAP partnership and offers certified infrastructure for HANA workloads. For organizations where the data and AI use case is central to the SAP investment thesis, Google Cloud deserves serious consideration.

The consideration: Google Cloud's enterprise sales and support organization is less mature than AWS or Azure, and its partner ecosystem for SAP implementations is narrower. Organizations should factor this into their evaluation.

Rather than starting with the hyperscaler and working backward, we recommend starting with four questions:

1. What are our existing enterprise cloud commitments? Most large organizations have negotiated enterprise agreements with one or more hyperscalers. Understanding the commercial implications of running SAP workloads within versus outside those agreements is the first step. In many cases, the commercial analysis alone significantly narrows the field.

2. What is our broader data and AI strategy? The hyperscaler that hosts SAP will also be the natural home for SAP data in analytics and AI use cases. If your organization has a clear view of the AI capabilities it wants to build on top of SAP data, that should inform the infrastructure decision.

3. What does our RISE with SAP evaluation look like? If RISE is under consideration, the hyperscaler options available under RISE — and the commercial terms associated with each — need to be evaluated alongside the direct hyperscaler options. In some cases, the RISE commercial structure makes one hyperscaler significantly more attractive than the others.

4. What are our regulatory and data residency requirements? For organizations in regulated industries or jurisdictions with data residency requirements, the availability of certified regions and the hyperscaler's compliance posture in relevant markets is a constraint that must be evaluated early.

The hyperscaler vendors, SAP, and system integrators all have commercial interests in the infrastructure decision. AWS, Azure, and Google Cloud each have incentives to position their platform as the right choice. SAP has its own preferences, which are reflected in the RISE commercial structure. System integrators often have preferred hyperscaler relationships that influence their recommendations.

Independent advisory — from an advisor with no commercial relationship with any of these parties — is the most reliable way to ensure that the infrastructure decision is made on the merits of your organization's specific situation, not on the basis of vendor incentives.

That is the analysis Astraeus Advisory Group provides.

To discuss your SAP infrastructure strategy with an independent advisor, contact a partner directly. No salespeople. No intake queues.