IBM InfoSphere DataStage v8.5 Sample Questions:

1. How are Shared Table definitions created from the DataStage Client?

A) No need to, DataStage table definitions are automatically saved as shared metadata.
B) Using the "Save" button on the columns tab on the Stage/Link editor
C) This cannot be done from the DataStage Client
D) Using the "Save" button on the "Table Properties" panel
E) Using the "Shared Table Creation Wizard" from the DataStage Client.

2. You are required to use a Make Vector stage in your job. What three requirements must be met in order to use this stage? (Choose three.)

A) All columns are combined into a vector of the same length as the number of columns.
B) Columns must be in consecutive order.
C) Input columns must form a numeric sequence.
D) Input columns must all be of the same data type.
E) Input columns that do not have a name of form <column_name>0 to <column_name>n, where <column_name> is the column name and 0 and n are the first and last of its consecutive numbers will not be included in the vector and will be dropped.

3. The given job processes large volumes of data using default partitioning, with sort/partition insertion enabled. Both source Data Sets are partitioned and sorted on the merge key. Assuming that the business logic allows all of the following actions, which design change would improve the performance of this job?

A) Before each input to the Merge stage, insert a Sort stage with "Don't Group, Previously Grouped" set on the merge key. Specify SAME partitioning on all inputs.
B) Set the buffer mode to "no buffer" on the output links of both Data Sets.
C) Use the Data Set Management Utility to combine both input Data Sets into a single source Data Set.
D) Set the "Preserve Partitioning" option on both Data Sets and replace the Merge stage with a Sort Funnel stage.

4. You write a job control routine to control a sequence of jobs running as a single unit of work. What are three valid job status types you can trap for? (Choose three.)

A) DSJS.RUNOK
B) DSJS.RUNFAILED
C) DSJS.RUNERROR
D) DSJS.START
E) DSJS.STOPPED

5. Your job is using the DB2 Enterprise stage as a target. In this job you are inserting data into a single existing table on a remote database. You are using Run-Time Column Propagation (RCP) on all of the links.
Which two job design configuration actions would you consider to effectively use RCP? (Select two)

A) Use a Modify stage prior to the DB2 Enterprise stage in order to drop the columns not found in the table.
B) In the DB2 Enterprise stage use Connection properties to manage column mappings.
C) Use DB2 partitioning properties on the input link to the DB2 Enterprise stage to control column layout in partitions.
D) In the DB2 Enterprise stage set the option named "Silently Drop Columns Not in Table" to "True".

Solutions: