Co-Authored by Suhail Syed
The growing popularity of container technology for enterprise software development is no secret. SAP realized this early on and, in the fall of 2017, released SAP Data Hub, a microservices-based platform that enables enterprises to orchestrate, aggregate, visualize, and generate data insights from across an entire business data landscape, such as Hadoop, Amazon S3, SAP HANA, and SAP Business Suite.
A response to growing complexity
In effect, SAP Data Hub is SAP’s response to growing complexity in the enterprise data landscape. For example, there is often a missing link between enterprise data—typically found in transactional systems—and big data such as that streamed by IoT devices. In addition, these data sources often collect in dispersed and siloed data lakes across an enterprise and various clouds. Compounding this challenge, organization must address compliance concerns, which include the need for security and data visibility.
For orchestration and management of containers, Kubernetes has emerged as a popular option. In fact, wherever Kubernetes is running, SAP Data Hub runs as well. Unfortunately, deployment and day-to-day on-premises Kubernetes operations remain challenging, especially for hybrid and multicloud deployments.
To help its customers meet this challenge, SAP sought a partner with whom it could work closely to deliver a turnkey, on-premises solution that would accelerate and facilitate SAP Data Hub operations. To meet those needs, SAP looked to Cisco and, specifically, to the Cisco Container Platform (CCP).
Responding to deployment challenges
Let’s look at some of the challenges SAP customers have had deploying SAP Data Hub and how CCP can make life easier for them. (See Figure 1.)
Figure 1 Areas Where Deploying SAP Data Hub Can Be a Challenge
To begin with, enterprises often want to store business-critical data on premises in their own data centers. This is especially true when discussing SAP systems; applications such as ERP and S4/HANA contain business-critical processes and data that organizations want to secure.
That said, there are many cloud services that enterprises might want to leverage, such as machine learning, advanced analytics, and cloud storage. So the question becomes how can they combine these on-premises and cloud systems? (See Figure 2)
Figure 2 On-Premises and Cloud Systems
Actually, running SAP Data Hub in the cloud is fairly straightforward, because all cloud providers have fully managed Kubernetes cloud services. However, the first challenge appears when deploying and operating in a hybrid environment. For example, you need to answer questions like “What flavor of Kubernetes do I install, how will I manage it, and who will support it?” (See Figure 3.)
Figure 3 Running SAP Data Hub in a hybrid environment raises questions
Additional aspects to consider when installing Kubernetes across multiple clouds appear in the following diagram. (See Figure 4.)
Figure 4 Additional Aspects to Consider When Installing Kubernetes
The second challenge arises when managing connectivity between SAP Data Hub running on premises and in the cloud, so the two systems can communicate with each other. For example, even though the on-premises environment can call into the cloud under some circumstances, typically a cloud service can’t call directly into the customer data center. (See Figure 5.) So, organizations need an efficient way to connect the cloud and on-premises environments.
Figure 5 Managing Connectivity between SAP Data Hub on Premises and in the Cloud Can Be Challenging
The third challenge is missing user management integration. If you’re operating two kinds of Kubernetes cloud services—one in the cloud and one on premises—then you need to replicate user access and authorization in both environments. And that gets complicated, as you can see in Figure 6.
Figure 6 Replicating User Access and Authorization in the Cloud and On Premises
To help SAP Data Hub users meet these challenges, SAP partnered with Cisco to integrate SAP Data Hub with CCP as an on-premises solution stack. (See Figure 7.) Organizations can now run SAP Data Hub on CCP, confident that they are using the best of open-source Kubernetes secured and supported on validated hardware by the industry-leading Cisco Technical Assistance Center (TAC) as their single point of contact. Note that in the hybrid cloud, organizations still run SAP Data Hub on a cloud provider Kubernetes platform, such as the Amazon Elastic Kubernetes Service (Amazon EKS).
Figure 7 Integrating SAP Data Hub with Cisco Container Platform in an On-Premises Solution Stack
Enterprises then use the Cisco Cloud Services Router (Cisco CSR) connection for a native, secure VPN-encrypted connection between the cloud and their premises, which enables communications between both environments. Cisco CSR is a software router that enterprises and cloud providers can deploy as a virtual machine in a provider-hosted cloud or in its own virtual environment. (See Figure 8.)
Figure 8 Cisco Cloud Services Router Connects the Cloud and a Customer Data Center
Finally, an organization can install, integrate, and manage a cloud provider’s Kubernetes service directly from CCP running on premises—and that includes robust user management. This capability makes it possible to manage multiple Kubernetes environments and provide security and compliance on a single platform. (See Figure 9.)
Figure 9 Providing Security and Compliance Through a Single Management Management Platform
Ultimately, this capability allows an organization to virtually span an SAP Data Hub that runs as a hybrid data processing platform, bridging cloud and premises, as you see in Figure 10.
Figure 10 Virtually Spanning an SAP Data Hub that Runs as a Hybrid Data Processing Platform
An example use case
A use case might include initially processing, aggregating, and filtering IOT data in the cloud. Once this process is finished, the data is sent on premises in a compressed—to reduce data egress costs—anonymized format that meets regulatory compliance requirements. The data is then retransformed efficiently and in a natively connected way to enrich relevant data warehouses using SAP Data Hub.
Simplifying, integrating, and centralizing
Ultimately, the combination of SAP Data Hub and Cisco Container Platform simplifies complex data landscapes. This solution allows enterprises to accelerate development because they can get data quickly to where it must be consumed. Multiple data sources can also be easily integrated, with the objective of developing valuable business insights more quickly. Finally, centralizing data management improves visibility and data governance.
Cisco and SAP teamed to integrate SAP Data Hub with Cisco Container Platform on Cisco HyperFlex and FlexPod, helping ensure that this containerized solution runs on high- performance, hyperconverged and converged hardware. By the way, SAP and Cisco are also working to do the same for SAP Data Intelligence—including graphics processing unit (GPU) capabilities to enable high-performance machine learning.
This integration offers a turnkey, security-hardened, end-to-end enterprise-grade solution with a choice of persistent storage options that lets SAP Data Hub users run production-grade Kubernetes environments easily and safely on premises, with the option to run them in hybrid cloud mode.
So if you’re looking to SAP Data Hub or SAP Data Intelligence to help simplify your complex enterprise data landscape, it’s worth your time to investigate how Cisco Container Platform can help you more easily deploy and manage Kubernetes environments, especially in hybrid and multicloud deployments.
To learn more about simplifying your data environment, visit Cisco, SAP, and Kubernetes container orchestration.
To learn more about how Cisco and SAP are working together, visit https://www.cisco.com/go/sap.