SAP HANA is the in-memory database platform at the heart of SAP S/4HANA, and HANA skills remain among the most in-demand in the SAP job market. Whether you are a developer, modeler, or Basis-adjacent consultant, interviewers will test your understanding of HANA’s architecture, modeling tools, and development techniques.
This guide covers 30 essential SAP HANA interview questions with clear, practical answers — from in-memory fundamentals to calculation views, CDS, AMDP, and code pushdown.
Architecture and Fundamentals
Q1. What is SAP HANA?
Answer: SAP HANA (High-Performance Analytic Appliance) is SAP’s in-memory database platform. It stores data primarily in RAM rather than on disk, which allows dramatically faster data processing — especially for analytics and aggregations. It combines transactional (OLTP) and analytical (OLAP) processing in a single system and serves as the database for SAP S/4HANA.
Q2. What does “in-memory” mean, and why does it matter?
Answer: In-memory means the database keeps its primary data copy in main memory (RAM) instead of reading from disk for every query. RAM access is orders of magnitude faster than disk access, so queries, aggregations, and calculations run much faster. HANA still persists data to disk for durability (through savepoints and logs), but the working copy lives in memory.
Q3. What is the difference between row store and column store in HANA?
Answer: In a row store, all fields of a record are stored together — efficient when you read entire records (typical for transactional operations). In a column store, each column is stored separately — efficient for analytics, because queries that aggregate a few columns (e.g., SUM of sales) only read those columns. HANA uses the column store by default for most application tables, since analytics is its strength, and the row store for specific system or transactional tables.
Q4. What is data compression in HANA, and why is it important?
Answer: Because memory is more expensive than disk, HANA compresses column-store data aggressively (using techniques like dictionary encoding and run-length encoding). Columnar data compresses very well since values in a single column tend to repeat. This keeps the memory footprint manageable despite the in-memory architecture.
Q5. What is the persistence layer in HANA?
Answer: The persistence layer writes data to disk to protect against power loss or crashes. It consists of data volumes (savepoints — periodic snapshots of memory to disk) and log volumes (redo logs recording every change). On restart, HANA reloads the last savepoint and replays the logs to recover.
Modeling: Views and CDS
Q6. What is a calculation view in SAP HANA?
Answer: A calculation view is a virtual data model that combines tables and other views using joins, unions, projections, and aggregations. It does not store data itself — it defines a query that executes when the view is consumed. Calculation views are the primary modeling artifact for analytics in HANA and come in graphical and SQL-script flavors.
Q7. What are the types of calculation views?
Answer: The main types are cube (with measures, for multidimensional analysis — historically “aggregation” level views) and dimension (master data, no measures). They can be built graphically or defined with SQLScript. Modern development favors the SQL-based (code-based) calculation views managed in development tools.
Q8. What is an analytic view vs an attribute view? (Classic HANA modeling)
Answer: In classic HANA modeling (pre-XSA, HANA 1.0 studio-based): attribute views modeled master/dimension data (like customer or product masters), analytic views modeled fact data with measures (like sales amounts) joined to attribute views, and calculation views combined everything with advanced logic. In current HANA versions, calculation views have largely absorbed these roles.
Q9. What is CDS in the HANA context?
Answer: CDS (Core Data Services) is SAP’s data modeling language for defining semantically rich data models. In the HANA/ABAP world, ABAP CDS views are defined in the ABAP layer but can push processing down to the HANA database. HANA CDS (in native HANA development) defines database-level artifacts. Both let you model entities, associations, and annotations in a declarative way.
Q10. What are annotations in CDS views?
Answer: Annotations are metadata attached to CDS entities and fields using the @ syntax. They control behavior and semantics — for example, @EndUserText.label provides a human-readable label, @Semantics.amount.currencyCode links an amount to its currency field, and UI annotations drive Fiori element rendering. Annotations separate what the data means from how it is displayed or consumed.
Q11. What is an association in a CDS view?
Answer: An association defines a relationship between CDS entities (e.g., sales orders to customers) without an explicit JOIN in the view definition. The join is generated only when the associated fields are actually used (on-demand join), which keeps queries efficient. Associations are a core reason CDS is preferred over classic database views.
Development: SQLScript, AMDP, and Code Pushdown
Q12. What is SQLScript?
Answer: SQLScript is HANA’s extension of SQL for writing database procedures and functions. It supports declarative set-based logic as well as imperative constructs (loops, conditionals). It is used in calculation views (SQL-based), stored procedures, and table functions — anywhere you need logic to run inside the database.
Q13. What is code pushdown, and why is it important?
Answer: Code pushdown means moving data-intensive logic from the application server down into the HANA database, so processing happens where the data lives. Instead of fetching millions of rows to the application server and looping over them in ABAP, you express the logic in SQL/CDS/AMDP and let HANA’s columnar engine process it in memory. This is the single most important performance principle in HANA-based ABAP development.
Q14. What is AMDP?
Answer: AMDP (ABAP Managed Database Procedure) is a framework that lets ABAP developers write database procedures (in SQLScript) that are managed as ABAP repository objects. You define a class method marked for AMDP, write the SQLScript inside, and call it from ABAP like a normal method. It is the standard way to implement logic that must run inside HANA when CDS alone is not expressive enough.
Q15. When would you use AMDP instead of a CDS view?
Answer: Use CDS views for declarative modeling (joins, associations, aggregations) — they are the first choice. Reach for AMDP when you need procedural logic that CDS cannot express: complex loops, multi-step transformations, dynamic SQL, or algorithms that need imperative code. Even then, keep the AMDP focused and push set-based parts into SQL.
Q16. What is a table function in ABAP CDS?
Answer: A CDS table function is a CDS view whose implementation is provided by an AMDP method. It looks like a normal CDS view to consumers (you can SELECT from it), but the logic inside is SQLScript code. This bridges the gap: full procedural power with a clean, consumable CDS interface.
Data Provisioning and Integration
Q17. What is SLT?
Answer: SLT (SAP Landscape Transformation Replication Server) is SAP’s real-time data replication tool. It captures database changes from a source system via triggers and replicates them to HANA (or other targets) with very low latency. It is commonly used to feed HANA sidecar scenarios or replicate ECC data into HANA-based systems.
Q18. What are the alternatives to SLT for loading data into HANA?
Answer: Common options include SDI (Smart Data Integration) for batch and real-time ETL-style flows, SDA (Smart Data Access) for virtual access to remote data without replication, flat-file uploads, and direct extraction tools. The right choice depends on latency requirements, data volume, and whether you want replication or virtualization.
Q19. What is the difference between SDA and SDI?
Answer: SDA (Smart Data Access) creates virtual tables pointing at remote data sources — queries federate out to the source at runtime, with no data copied. SDI (Smart Data Integration) physically moves/transforms data into HANA via flowgraphs and adapters. Use SDA when you want live access without duplication; use SDI when you need persisted, transformed data in HANA.
Tools and Administration Basics
Q20. What tools are used for HANA development and modeling?
Answer: Modern HANA development uses the ABAP Development Tools (ADT) in Eclipse for ABAP CDS and AMDP, and SAP Business Application Studio or VS Code with SAP extensions for native HANA (CAP/XSA) development. The older HANA Studio was the classic Eclipse-based tool for HANA 1.0-era modeling and administration.
Q21. What is a delivery unit (DU)?
Answer: In classic HANA (1.0), a delivery unit was the transport package for HANA repository content (models, procedures). In current versions, content is managed through standard ABAP transports or HANA deployment infrastructure (HDI) containers, depending on the development model.
Q22. What is partitioning in HANA, and when is it used?
Answer: Partitioning splits a large column-store table across multiple hosts or cores so queries can run in parallel. It is used for very large tables to improve query performance and manage memory. Common schemes include hash, range, and round-robin partitioning, chosen based on query patterns.
Q23. What is delta merge in HANA?
Answer: Column-store tables in HANA have a main store (compressed, read-optimized) and a delta store (write-optimized, uncompressed). New writes go to the delta store; periodically, a delta merge moves delta data into the main store and recompresses it. Frequent merges keep read performance high; the merge can be triggered manually or automatically.
S/4HANA Context
Q24. How does HANA relate to S/4HANA?
Answer: SAP S/4HANA is SAP’s next-generation ERP suite, and it runs only on the SAP HANA database. S/4HANA’s simplified data model (fewer aggregate tables, longer key fields like MATNR) is designed around HANA’s in-memory, columnar capabilities — the database is not just storage, it is part of the application architecture.
Q25. What is the difference between HANA as a database and S/4HANA as an application?
Answer: SAP HANA is the database platform (in-memory, columnar). SAP S/4HANA is the ERP application suite that runs on it. You can use HANA as a standalone database for custom applications, but S/4HANA requires HANA — the two are often confused because S/4HANA cannot run on any other database.
Q26. What are CDS views used for in S/4HANA?
Answer: In S/4HANA, ABAP CDS views are the standard data modeling and consumption layer. They power Fiori apps (via OData), embedded analytics, and custom reports. SAP ships thousands of standard CDS views, and custom development is expected to build on CDS rather than direct table access.
Performance and Best Practices
Q27. What are the golden rules of HANA performance?
Answer: The essentials: push code down to the database (filter and aggregate in SQL/CDS, not in ABAP loops); select only needed columns (columnar reads are cheap only for the columns you touch); avoid row-by-row processing of large datasets in the application layer; and mind the delta merge and partitioning for very large tables.
Q28. What is a common HANA performance anti-pattern in ABAP?
Answer: Selecting a huge unfiltered dataset into an internal table and then looping with nested SELECTs or filtering in ABAP. The HANA way is to express the filter, join, and aggregation in a CDS view or AMDP so the database returns only the final result set.
Q29. How do you debug performance issues in HANA queries?
Answer: Start with the SQL plan (EXPLAIN PLAN) to see how HANA executes the query, check which operations dominate (column scans, joins, aggregations), and look at tools like the SQL Plan Cache and performance traces in the DBA cockpit or HANA cockpit. Often the fix is rewriting the logic to be more set-based or adding appropriate filters.
Q30. What should a HANA beginner learn first?
Answer: Start with the fundamentals: in-memory concepts, row vs column store, and basic SQL. Then learn calculation views and CDS modeling, then code pushdown principles with AMDP. Hands-on practice in a trial HANA environment matters more than theory — build views, write SQLScript, and measure the difference pushdown makes.
Frequently Asked Questions
Q1. Is SAP HANA a database or an application?
SAP HANA is a database platform. SAP S/4HANA is the ERP application that runs on it. The names are related but refer to different layers of the stack.
Q2. Do I need to learn SQLScript to work with HANA?
For modeling and analytics work, graphical/SQL calculation views and CDS cover a lot. For advanced development — complex transformations, table functions, performance-critical logic — SQLScript (usually via AMDP) becomes important.
Q3. What is the difference between HANA 1.0 and HANA 2.0 modeling?
HANA 1.0 used the repository-based classic modeling (attribute/analytic/calculation views in HANA Studio). HANA 2.0 moved toward code-based development: SQL-scripted calculation views, HDI containers, and development in modern IDEs. Calculation views remain central in both.
Q4. Can I practice HANA without a paid system?
SAP offers trial and developer editions of HANA (such as the HANA Express edition) that you can run locally or in the cloud for learning. These are the best way to get hands-on experience.
Q5. How is HANA different from a traditional database like Oracle?
The core difference is architectural: HANA is in-memory and column-oriented by design, built for combined transactional and analytical workloads. Traditional databases are disk-based and row-oriented, with analytics typically offloaded to a separate data warehouse.
Q6. Will learning HANA help my SAP career?
Yes. HANA underpins S/4HANA, which is where the SAP market is heading. HANA modeling, CDS, and code-pushdown skills are consistently in demand for S/4HANA implementation and support roles.
The Bottom Line
SAP HANA interviews test whether you understand why HANA is different — in-memory processing, columnar storage, and the discipline of pushing logic down to the database — not just its vocabulary. Master the fundamentals (row vs column store, delta merge, partitioning), the modeling tools (calculation views, CDS, associations, annotations), and the development techniques (SQLScript, AMDP, code pushdown), and you will handle the vast majority of HANA interview questions with confidence.
Preparing for a functional interview too? See our Top 40 SAP FICO Interview Questions and Answers for the finance side.