SAP Cloud Integration Cookbook (Martin Koch, Thorsten Reisinger etc.)（Z-Library）

# SAP Cloud Integration Cookbook — Reading Guide ## 【One-Line Pitch】 A practical, recipe-based guide for integration developers who need to master SAP Cloud Integration within SAP Integration Suite—covering everything from core architecture and mapping techniques to security, transport management, and real-world hybrid integration scenarios. Ideal for SAP PI/PO veterans transitioning to the cloud and for new integration developers building scalable iFlows in hybrid landscapes. --- ## 【Book Arc】 - **Opening (~0%–9%)**: Introduces Cloud Integration as the core capability of SAP Integration Suite, contrasting it with classic SAP PI/PO. Covers the "clean core" principle for cloud-enabled SAP landscapes, then walks through the key tenant components—design time, runtime engine, edge integration cell, keystore, monitoring, and partner directory—explaining how they work together to model, execute, and manage iFlows. - **Early (~9%–28%)**: Dives into runtime behavior and infrastructure details. Explains quality of service (QoS) levels—At Least Once, Best Effort, Exactly Once—and how they control message reliability. Introduces the edge integration cell for private on-premise data exchange via Kubernetes, monitoring capabilities (including API access for Splunk/Cloud ALM integration), and the partner directory as a central repository for B2B partner metadata with dynamic runtime parameter loading. - **Early–Middle (~28%–38%)**: Systematically catalogs Cloud Integration capabilities by function: message transformation (graphical mapping, Groovy/JavaScript scripting, XSLT 2.0), routing (router, multicast, splitter, join/gather), and data persistence (persist message, data store operations with the 32 GB SAP ASE limit, write variables). Also covers idempotency concepts and the idempotent process call for avoiding duplicate executions. - **Middle (~38%–47%)**: Focuses on message protection and security mechanisms (PGP, PKCS #7 encryption), then shifts to prepackaged content from the SAP Business Accelerator Hub and community packages under Apache 2.0. Explains the crucial distinction between unmodified use (metered but not billed) and modified use (billed), plus detailed message metering rules—250 KB per message unit, splitter counting logic, and retry handling. - **Middle–Late (~47%–53%)**: Covers operational and governance topics: transport management, logging strategies, and the architectural principles that tie everything together—technical security via TLS/SAML/role models, organizational separation through role-based access, and scalability through modular capabilities. The excerpts suggest the book closes with practical interface scenarios that apply the accumulated techniques. --- ## 【Key Takeaways】 - **Clean core is non-negotiable in cloud SAP landscapes** (Early): Direct modifications to SAP standard—custom ABAP in standard transactions or database table changes—jeopardize automatic updates in RISE with SAP models. Use key user tools, in-app extensions, or side-by-side BTP developments with official APIs, events, and CDS views instead. - **Cloud Integration design time differs fundamentally from SAP PO** (Early): Unlike PO's ESR-based, strictly typed metadata approach, Cloud Integration uses a browser-based low-code/pro-code environment with drag-and-drop modeling, integrated Groovy/JavaScript scripting, and package-based versioning. This reflects a shift from centralized governance to agile, continuous delivery. - **QoS levels give you differentiated control over message reliability** (Early): At Least Once (default, processes duplicates), Best Effort (synchronous, no retries, for real-time noncritical scenarios), and Exactly Once (deduplicates via message IDs). Choose based on business criticality and fault-tolerance requirements. - **The partner directory enables massive B2B scalability** (Early): Store partner-specific values (URLs, IDs, authentication data) centrally and load them dynamically at runtime based on message context. One generic iFlow can serve hundreds or thousands of partners with zero downtime for new connections—no redeployment needed. - **Idempotency prevents duplicate processing disasters** (Early–Middle): An operation is idempotent if calling it once or multiple times yields the same result. The idempotent process call checks message IDs to discard or mark duplicates, which is critical for handling technical retries and network problems. - **Message metering has specific rules that affect cost** (Middle): Billing counts outgoing messages via receiver adapters, not iFlow calls. Every 250 KB (or portion) counts as one message; splitters count each split individually (highest-count splitter wins); retries count separately; JMS and process direct adapters are free. Currently (as of August 2025), each iFlow run with a receiver adapter counts as one message regardless of size. - **Prepackaged content has a billing trap** (Middle): Using iFlows from official packages unmodified means messages are metered but not billed. Once you edit them—adding logic, transformations, or security—you're billed per message. Check SAP Note 2942344 for details. - **The data store has hard limits** (Middle): Persist message and data store operations share the tenant database with a 32 GB overall space limit (SAP ASE). Use data stores for push-pull patterns and cross-iFlow message sharing, but plan storage carefully. --- ## 【Reading Tips】 - **Skim the opening architecture chapters (0%–9%)** if you're already familiar with SAP Integration Suite basics; they're essential for newcomers but largely conceptual. Focus on the design time versus SAP PO comparison—it frames everything that follows. - **Deep-read the capabilities catalog (~28%–38%)**: This is the reference core of the book. Pay special attention to the routing and persistence sections—they're the building blocks for most real-world integration patterns. The idempotency explanation here is worth memorizing. - **The metering section (~44%–47%) deserves careful study**: The billing rules are counterintuitive (splitter counting, 250 KB rounding, receiver-adapter focus). Work through the numerical examples yourself—they're the kind of detail that saves you from costly surprises in production. - **Treat this as a cookbook, not a novel**: Skip between chapters based on your current integration problem. The "recipes" format means each section stands alone, so use the table of contents to jump to adapters, mapping, or security as needed. - **If you're a SAP PO veteran, focus on the paradigm shifts**: The book repeatedly contrasts PO's structured ESR approach with Cloud Integration's flexible, package-based model. Understanding these philosophical differences matters more than memorizing individual features. --- ## 【Coverage Limits】 The excerpts cover roughly the first half of the book (through ~53%), focusing on architecture, capabilities, and metering. Detailed recipe content on specific adapters, Groovy/XSLT mapping techniques, transport management, and the promised practical interface scenarios are not covered in this guide—the excerpts confirm these topics exist but don't detail them. --- ##