IBM InfoSphere DataStage v8.5 認定 C2090-421 試験問題:

1. A 100MB input dataset has even distribution across 400 unique key values.
When run with a 4-node configuration file, which two changes could improve sort performance in this scenario? (Select two)

A) Set $APT_TSORT_STRESS_BLOCKSIZE to 50M
B) Specify Round Robin partitioning on the input link to the Sort stage.
C) Specify "Restrict Memory Usage" to 60MB on the Sort stage properties.
D) Convert all numeric fields to varchars.
E) Alter the Sort stage properties to perform a stable sort.

2. You are asked by management to document all jobs written to make future maintenance easier. Which two statements are true about annotations? (Choose two.)

A) The Description Annotation stage contains both the short and full descriptions for the job.
B) The background for the Description Annotation stage can be changed for each unique stage.
C) The full job description can be identified within the Description Annotation stage.
D) The Description Annotation stage can be added several times at different locations to identify business logic.

3. In the job fragment shown in the exhibit, there are two Sort stages: the first (Sort1) sorts the data by state, the second (Sort2) sorts the data by state and area, No repartitioning occurs between the stages.
You have been asked to reduce the number of sorts to improve the performance of the job.

Which two ways might be used to accomplish this without changing how the job functions? (Select two)

A) In the second Sort stage change the Sort Mode property of the area column to "Don't sort, previously sorted.".
B) Remove the second Sort stage. Add area as an additional sort key.
C) In the second Sort stage change the Sort Mode property of the state column to "Don't sort, previously sorted.".
D) Remove the first Sort stage. Retain the existing sort keys in the second Sort stage.

4. What three element Names are required to create Shared metadata? (Choose three.)

A) Schema
B) Host System
C) Database Instance
D) Database
E) Project Name

5. You are setting up the parallel engine configuration file. Which two statements are true about file system efficiency? (Select two)

A) Using multiple scratch spaces in a logical node improves performance.
B) In a clustered environment where NAS is used for a common storage, specify the internal disk for scratch disk.
C) Creating scratch disk space on an internal disk improves sort performance than NFS drive.
D) Specify a directory for the resource disk in the partition where the engine is installed.

質問と回答：