SAP BTP Runtimes Comparison Reference
Overview
SAP BTP provides flexible runtime options allowing development teams to select environments based on technical requirements, developer skills, and cost considerations.
Runtime Options
1. Cloud Foundry Runtime
Best For: Traditional PaaS applications, rapid deployment
Characteristics:
- Managed application runtime
- Buildpack-based deployments
- Simple scaling with
cf scale - Memory-based billing
Languages:
- Node.js
- Java
- Python
- Go
- PHP
Use Cases:
- Cloud-native applications
- Microservices
- API backends
- Scheduled jobs
2. Kyma Runtime
Best For: Kubernetes-native applications, containerized workloads
Characteristics:
- Managed Kubernetes cluster
- Istio service mesh included
- Helm-based deployments
- Container-based billing
Technologies:
- Kubernetes
- Docker containers
- Helm charts
- Istio service mesh
- Serverless functions
Use Cases:
- Container-based applications
- Microservices with service mesh
- Event-driven architectures
- Custom runtime requirements
3. ABAP Environment
Best For: Organizations with ABAP expertise, SAP extensions
Characteristics:
- Managed ABAP platform
- RAP programming model
- Integrated SAP Fiori Launchpad
- ACU/HCU-based billing
Languages:
- ABAP (restricted cloud-safe subset)
Use Cases:
- SAP S/4HANA extensions
- Enterprise business applications
- Migrating ABAP custom code
- ISV solutions
Comparison Matrix
| Aspect | Cloud Foundry | Kyma | ABAP Environment |
|---|---|---|---|
| Programming | Node.js, Java, Python | Any containerized | ABAP Cloud |
| Deployment | cf push / MTA | Helm / kubectl | gCTS / Landscape Portal |
| Scaling | Horizontal (instances) | Kubernetes pods | Elastic ACUs |
| Persistence | HANA Cloud, PostgreSQL | HANA Cloud, any | HANA Cloud (managed) |
| UI | SAPUI5, Fiori Elements | Any | Fiori Elements |
| Billing Model | Memory consumption | Cluster + resources | ACU + HCU |
Metering Details
Cloud Foundry
- Start: When runtime memory is consumed (e.g., application deployment)
- Metric: GB-hours of memory
- Optimization: Stop unused applications, right-size memory
Kyma
- Start: When Kyma runtime is enabled (cluster creation)
- Metric: Dedicated Kubernetes cluster resources
- Optimization: Node autoscaling, resource limits
ABAP Environment
- Start: When ABAP system is provisioned
- Metric: ABAP Compute Units (ACU) + HANA Compute Units (HCU)
- Optimization: System hibernation, elastic scaling
Choosing a Runtime
Choose Cloud Foundry When:
- Developing CAP applications with Node.js or Java
- Need simple deployment model
- Want buildpack-managed dependencies
- Cost-sensitive workloads
Choose Kyma When:
- Need Kubernetes-native features
- Require service mesh (Istio)
- Building container-based microservices
- Need serverless functions
- Custom runtime requirements
Choose ABAP Environment When:
- Have existing ABAP expertise
- Extending SAP S/4HANA
- Need tight SAP integration
- Building partner add-on products
- Migrating custom ABAP code
CAP Support Across Runtimes
Cloud Foundry
# Deploy CAP to Cloud Foundry
mbt build
cf deploy mta_archives/my-app.mtarKyma
# Deploy CAP to Kyma
cds add helm
helm upgrade --install my-app ./chartABAP (Not CAP)
- Use RAP (ABAP RESTful Application Programming Model)
- Different programming model, similar concepts
Runtime-Specific Documentation
Source Documentation
- Runtime Benefits: https://github.com/SAP-docs/btp-developer-guide/blob/main/docs/strengths-and-benefits-of-the-runtimes-and-their-programming-models-86688d1.md
- CF/Kyma with CAP: https://github.com/SAP-docs/btp-developer-guide/blob/main/docs/sap-btp-cloud-foundry-and-sap-btp-kyma-runtimes-with-cap-0f9cfe9.md
- ABAP Environment: https://github.com/SAP-docs/btp-developer-guide/blob/main/docs/sap-btp-abap-environment-with-abap-cloud-174b229.md