000-418 試験問題を無料オンラインアクセス

When tuning a parallel process, it is necessary to measure the amount of system resources that are used by each instance of a stage. Which two methods enable the collection of CPU time used by each instance of a stage? (Choose two.)

• A.Invoke vmstat before the job run and after the job completes.
• B.Set the environment variable $DS_MAKE_JOB_REPORT=2.
• C.Select the Record job performance data check box from Job Properties.
• D.Set the environment variable $APT_PM_PLAYER_TIMING=true.

What is the lowest CPU cost partitioning method for parallel stage to parallel stage?

• A.Modulus
• B.Same
• C.Entire
• D.Range

Which three statements are true about Transfomer stages? (Choose three.)

• A.Stage variables are executed before output link derivations.
• B.Constraints can keep a record from being written to an output link.
• C.Constraints are executed before output link derivations.
• D.Derivations are written in C++.
• E.@PARTNUM returns the number of job partitions currently active.

In which two situations is it appropriate to use a Sparse Lookup? (Choose two.)

• A.When the output of the Lookup stage needs to be hashed partitioned.
• B.When invoking a stored procedure within a database per row in the streaming link.
• C.When accessing DB2 data using the DB2 API stage.
• D.When reference data is significantly larger than the streaming data (100:1).

A DataStage job uses an Inner Join to combine data from two source parallel data sets that were written to disk in sort order based on the join key columns. Which two methods could be used to dramatically improve performance of this job? (Choose two.)

• A.Explicitly specify hash partitioning and sorting on each input to the Join stage.
• B.Unset the Preserve Partitioning flag on the output of each parallel data set.
• C.Add a parallel sort stage before each Join input, specifying the "Don't Sort, Previously Grouped" sort key mode for each key.
• D.Set the environment variable $APT_SORT_INSERTION_CHECK_ONLY.