Exam Certified Data Engineer Professional topic 1 question 31 discussion

Actual exam question from Databricks's Certified Data Engineer Professional

Question #: 31
Topic #: 1

[All Certified Data Engineer Professional Questions]

A junior data engineer is working to implement logic for a Lakehouse table named silver_device_recordings. The source data contains 100 unique fields in a highly nested JSON structure.
The silver_device_recordings table will be used downstream to power several production monitoring dashboards and a production model. At present, 45 of the 100 fields are being used in at least one of these applications.
The data engineer is trying to determine the best approach for dealing with schema declaration given the highly-nested structure of the data and the numerous fields.
Which of the following accurately presents information about Delta Lake and Databricks that may impact their decision-making process?

A. The Tungsten encoding used by Databricks is optimized for storing string data; newly-added native support for querying JSON strings means that string types are always most efficient.
B. Because Delta Lake uses Parquet for data storage, data types can be easily evolved by just modifying file footer information in place.
C. Human labor in writing code is the largest cost associated with data engineering workloads; as such, automating table declaration logic should be a priority in all migration workloads.
D. Because Databricks will infer schema using types that allow all observed data to be processed, setting types manually provides greater assurance of data quality enforcement.
E. Schema inference and evolution on Databricks ensure that inferred types will always accurately match the data types used by downstream systems.

Show Suggested Answer

Suggested Answer: D 🗳️

by hammer_1234_h at Sept. 14, 2023, 1:07 a.m.

Comments

Submit Cancel

RafaelCFC

Highly Voted 11 months, 3 weeks ago

Selected Answer: D

A is wrong, because Tungsten is a project around improving Spark's efficiency on memory and CPU usage; B is wrong because Parquet does not support file editing, it only supports overwrite and create operations by itself; C is wrong because completely automating schema declaration for tables will incur in reduced previsibility for data types and data quality; E is false because unlucky sampling can yield bad inferences by Spark;

upvoted 14 times

...

hal2401me

Highly Voted 9 months, 1 week ago

from my exam today, both C & D are no longer available, so they can't be correct. E & A are available. E states "always accurate" so I hesitate to choose it. There is a new option stating like "delta lake indexes first 32column in delta log for Z order and optimization"(not sure I remember exactly, it looks statementfully correct). and I chosed this "new" option. Because, this should impact the schema decision by putting high-usage field in the first 32 columns.

upvoted 7 times

...

Tedet

Most Recent 3 months, 3 weeks ago

Selected Answer: D

Explanation: Databricks can infer schema when reading data, but automatic schema inference doesn't always guarantee the accuracy of data types. For complex or highly-nested structures, schema inference might not always align with the actual data quality or the needs of downstream applications, and manual type definition ensures that the schema is more consistent and predictable. While automatic inference is useful for quick analysis or exploratory work, manual schema definition provides better data quality assurance in production workloads, especially when dealing with large, complex data structures. Conclusion: This statement correctly emphasizes the importance of manual schema declaration to ensure data quality enforcement and consistency, especially when dealing with complex structures. Best option.

upvoted 1 times

...