HANA EXAM 2024 Flashcards
Which of the following is true about column store tables?
They favour calculations on a few columns only
They favour table scans and / or aggregations based on the values of a few columns and many rows
A customer has a BW system with large daily loading of data into some of their BW systems. Which of the following statements is true in terms of the recommended merge configuration the customer should apply to the target tables?
Disable automerge for the target table and set smart merge to allow the application to suggest when a merge will take place
You create a column store table and need to load 20 million records to it. This will be done in one batch load. Automerge is disabled and smart merge is set on this table. When loading has finished and a merge takes place, how are the committed transactions handled?
Merged from existing delta to new main
With respect to the max_concurrency parameter, which of the following is not true?
Its value is the maximum amount of memory in GB that can be allocated to a process of a tenant database
You identify 10 tables that you will replicate to a HANA DB using SLT. Rather than just replicating like for like data, you need to apply some transformations. Where would this take place?
On the SLT system
You are monitoring a HANA DB following a customer’s go-live an there was a period of both high memory and CPU consumption between 15:47 and 16:08. to analyse you use the thread sampling scripts to try to identify the offending query. From the results of the script, which metric helps with identifying this query?
Statement Hash
During a TPO service, you run the mini checks script and notice that the check for average database request time exceeds the indicated KPI. After further analysis, you see that the elapsed time for most expensive statement on the system, the INSERT operation, is 98% lock wait time. Which of the following would make sense as a recommendation in this situation?
Check the application side if the critical time frame between the INSERT and the next COMMIT can be reduced
Check from an application and/or scheduling perspective if the number of concurrent changes of the same records by different transactions can be reduced
Within the context of Multitenant Database Containters, each database is self-contained and fully isolated with its own:
Persistence
Repository
Database catalog
Savepoints play an important role in a HANA DB as this is the method used to persist changes made on a given set of data. Which of the following statements are true about savepoints?
Savepoints have a default time interval of 5 minutes which can be configured
Savepoints contain undo logs that contain information about all open transactions
MVCC in row store is accomplished by using which two storages?
Persistent storage
Transaction version memory (TVM)
You create a column store table with 2 million records and execute a ‘SELECT*’ to load all columns into memory. You notice that the table size is larger in memory than the size it occupied on disk before you executed the query. What may be a reason for this?
Certain indexes related to the table only exist in memory
Just before a savepoint a table can be larger in memory as logs are not part of the table on disk
During a TPO service, you notice that the memory consumption on the HANA system is consistently high, and the customer often experiences out-of-memory situations. Which of the following makes sense to check for your root cause analysis?
Check the size of the column store
Check for any dominant heap allocators
