SAP HANA Interview Questions & Answers
SAP HANA Interview Questions
1. What led to the invention of SAP HANA technology?
Answer: The following led to the invention of SAP HANA technology:
- Information explosion where data was growing massively from Gigabytes to Terabytes to Petabytes and business required analytics based on this enormous data
- Instant data access anytime and anywhere became the customer need to make real time decisions
- Business required a flexible way to analyze current and historic information for various reporting solutions.
2. Which technologies did SAP HANA evolve from?
Answer: SAP HANA evolved by combining earlier developed technologies, BW Accelerator and Max DB with its in-memory capabilities.
3. Is SAP HANA a software or hardware?
Answer: SAP HANA is a combination of hardware and software; it is delivered as an optimized appliance in co-operation with SAP’s hardware partners for SAP HANA.
4. What are the main components in SAP HANA?
Answer: The main components of SAP HANA are:
- SAP In-Memory Database (IMDB)
- In-Memory Computing studio and
- Data replication components (SLT, BODS, etc.)
5. What is SAP HANA?
Answer: SAP HANA is an in-memory technology supported by column-based storage and high data compression that allows processing of massive volumes of data and high speed business reporting. It allows its customers to explore and analyze huge volumes of data from any data source in real time with unprecedented performance. In comparison to traditional RDBMS systems, it is much simpler and faster.
6. What are the capabilities and benefits that HANA offers?
- Real time data
- Faster queries on large volumes of data
- Flexible modeling
- Minimized data duplication
- No aggregate tables
7. What are the basic technology concepts in SAP HANA?
Answer: The basic technology concepts in SAP HANA are:
- In-Memory where data resides on main memory than on disk
- column based database, Data compression and pushing application logic to the database layer
- Parallel processing and multi-core CPUs leveraging the new hardware technology
- What is the benefit of In-Memory in SAP HANA?
Answer: The main benefit of using in-memory database is that accessing data from main memory is much faster than accessing on disk. A very high-speed bus connects the main memory directly to the processors, whereas in hard disks a chain of buses and controller are involved.
9. Why is SAP HANA fast?
Answer: SAP HANA is fast for the following reasons:
- HANA stores information in electronic memory as compared to regular RDBMS technologies that store information on hard disks
- Besides, most SAP systems have the database on one system and a calculation engine on another, and they pass information between them. With HANA, all this happens within the same machine.
10. What is columnar storage and how does it support faster access of data?
Answer: Columnar database stores data in a sequence of columns; the entries of a column get stored in contiguous memory locations. This phenomenon is called columnar storage.
Column store is optimized for READ; only the selected columns will be read during query processing, hence it performs well. It offers significant advantages of data compression or encoding data into fewer bits allowing larger volumes of data in main memory and higher performance in selection and aggregation queries.
11. Are column-based tables always better than row-based tables?
Answer: No. There are business cases where row based tables are advantages over column, like in frequently updated databases. If the database is frequently updated or inserted, row-based tables perform faster as they are optimized for write operations.
12. What is the difference between row store and column store?
Answer: The row store is optimized for WRITE operations and is easy to insert/update. All data has to be read during selection, even if only a few columns are involved in the selection process.
Compared to this, the column store is optimized for performance of READ operations and do not support easy insert/update. After selection, selected rows have to be reconstructed from column.
13. Can you have row store tables in SAP HANA?
Answer: HANA can have column or rows stores; there is no technical limitation. If you have row-store table in HANA you cannot create any column views on top those tables. Typically, metadata or rarely accesses data is stores in a row-store format.
14. How do you decide if the table should be row or column store in your project?
Answer: If you want to populate the tables with huge amounts of data that should be aggregated and analyzed fast and benefited from compression mechanisms, then column store is a better option. If you want to report on all the columns, then row store is more suitable.
Simple rule of thumb in HANA is use a column table unless specified.
15. How does insert or update work faster in HANA environment?
Answer: SAP HANA do not write directly into column store tables while inserting or updating the data as column store is not optimized for write operations. It first writes the data into a row store buffer which is write-optimized and hence faster. It then takes that data, restructures it and pushes it into a column-oriented store. As a final step, it pushes this restructures columnar data into the main column-oriented table. By following this process, SAP HANA makes use of the row store for write operations to ensure faster performance.
16. What degree of data compression is expected in SAP HANA?
Answer: The degree of data compression depends on the number of unique values in the data; the fewer the unique values, the better the data compression.
17. What is Delta Merge and how does it support faster read operations?
Answer: Delta merge moves the data from WRITE optimized Delta memory to READ optimized and compressed Main memory. It transforms the data into an optimized format in terms of memory consumption and read performance. By merging the data into the main storage which s column store, read operations would be faster.
18. What are the different ways of performing delta merge operation?
Answer: Delta merge can be done automatically by using smart merge technology or manually using MERGE DELTA OF function in SQL statement or using right click option in HANA studio.
19. When you run a query before delta merge, will you lose the data in the delta storage in the result set?
Answer: No. During any read operation data is always read from both main and delta storages and result set is merged.
20. What is memory latency and how does it hit the performance?
Answer: While executing any application logic on the data, the application has to get the data from the database, process it, and possibly send it back to the database to store the results. Sending data back and forth between the database and the application usually involves communication overhead and is limited by the speed and throughput of the network, this is memory latency.
21. How does SAP HANA handle the latency problem?
Answer: In SAP HANA, the calculations and application logic are done at the database level thereby reducing the overall processing time.
22. How does SAP HANA support parallel processing?
Answer: SAP HANA leverages the multi-core processors, multi-processor servers and scales out into a distributed landscape to support parallel processing.
23. Which are the top use cases in SAP HANA?
Answer: The top use cases of SAP HANA are:
- Real time financial planning
- Customer segmentation
- Genome analysis
- Profitability analysis and
- Detective HANA
24. What is the difference between SAP BWA and SAP HANA?
Answer: SAP BWA is only for SAP BW data, its main aim is to accelerate a portion of the BW data which is crucial for business reporting.
SAP HANA is much more than BWA, it replicates data from SAP ECC, BW or any other non-SAP source. Besides faster reporting, HANA supports flexible modeling and change management.
SAP HANA FAQs on Architecture:
25. What are the primary prerequisites for SAP BW on HANA?
Answer: The primary prerequisites for SAP BW on HANA are:
- Upgrade to SAP Net Weaver 7.02 or above
- Migrate database (RDBMS) to HANA DB
26. What is the Operating system requirement for SAP HANA?
Answer: SUSE Linux enterprise Server
27. Can HANA Server be configured via scale up or scale out configurations?
Answer: Yes, SAP HANA can be configures via Scale up and Scale out.
28. Name the servers operational in a HANA database.
Answer: The different servers operational in a HAN database are:
- Index Server
- Preprocessor Server
- Name Server
- Statistics Server
29. What is the role of each server in the HANA database?
- Index Server plays the prime role; it holds all the data and performs the query operations.
- Preprocessor Server processes the unstructured data and is typically used for Text Data Analysis.
- Name Server holds the landscape information. In a distributed system, name server contains the statuses of the active components and the data located on each server.
- Statistics server collects the information related to performance and resource consumption
30. What are the different services present in the HANA appliance?
Answer: The different services present in the HANA appliance are SAP Host Agent, Software Update Manager (SUM), SAP CAR, LM Structure.
31. What is the functionality of SUM and LM structures?
Answer: SUM allows automatic download and installation of SAP HAN versions and upgrades from SAP Marketplace.
LM structure holds details on the current product version installed.
32. Which component coordinates and tracks the database transactions?
Answer: Transaction Manager coordinates and the database transactions in SAP HANA.
33. How is an application query processes by the Index Server in SAP HANA?
Answer: The client requests in the application layer are passed down as SQL statements to the Request Processing and Execution Control components. The SQL Processor accepts the incoming SQL requests and executes the same as per the plan generated by the SQL Optimizer.
34. What is the role of MDX engine in the HANA server?
Answer: Multidimensional query requests from OLAP systems or analytical applications are processes by the MDX engine. Multidimensional data is stores in cubes and can only be queried using the MDX language.
35. Which are the two relational engines in In-Memory computing engine (IMCE)?
Answer: The two relational engines in IMCE are:
- Row store
- Column store
Both row store and column store are in-memory databases.
36. What are the key architecture points to be considered to ensure business continuity?
Answer: The key architectural points that ensure business continuity are:
- High availability per Data Center
- Disaster tolerance between Data Centers
37. How does HANA hardware support High Data Availability?
Answer: The Scale Out architecture with Standby node, delivered by the hardware partners Fujitsu, HP, IBM, etc. support high data availability in SAP HANA. Different servers are tightly connected to work together as a single system or cluster in this scale out approach and this improves performance and availability.
38. How is HANA hardware structures towards Disaster Tolerance?
Answer: Server clustering forming a distributed landscape ensures effective disaster tolerance in HANA environment. The nodes in a cluster would be stationed at different locations, thereby ensuring data availability in case of natural disasters or accidents at any particular region.
39. What is Scale-Out approach?
Answer: When the memory requirements go beyond a single server or to ensure high availability in cases of node failures, data is places across a group of servers as a distributed landscape; this is scale-out approach.
40. How is the Scale-Out architecture configured in HANA environment?
Answer: In the Scale-Out architecture where data is placed across a group of servers, N active servers and one standby server are configured for each cluster along with shared file system across all servers. During startup one server gets elected as active master. The active master assigns a volume to each starting index server or no volume in case of standby servers. This is to ensure high data availability and efficient disaster tolerance.
41. Which services will be active on each of the nodes in a Scale-out landscape?
Answer: Name server and index server will be active on all nodes; Statistics server will be active only on the active node and only Name server will be active on the standby node. During startup, one server in the cluster becomes the active master which assigns volumes to the remaining index servers.
42. How is master name server failure handled in the distributed landscape?
Answer: If the Name server in the master node fails, another of the remaining name-servers will become the active node.
43. How is Master Index server failure handled in the distributed landscape?
Answer: In case of Master Index Server failure, the Master Name server will be detect the Index server failure and triggers the failover. During this process, the master name server assigns the volume to the Standby server.
44. What is the significance of XS Engine (Extended Application Service)?
Answer: XS engine allows web-based applications to connect to SAP HANA database; clients can fetch data through HTTP connection. This is an optional component in SAP HANA appliance.
Enroll for Live Instructor Led Online SAP HANA Training