Cloudera CDP Data Engineer - Certification CDP-3002 Prüfungsfragen mit Lösungen:
1. You need to securely store sensitive data within your Spark application and access it only from authorized nodes. How can you leverage Cloudera security features to achieve this?
A) Use Cloudera Sentry for role-based access control and data masking
B) Implement custom encryption/decryption logic within your application
C) Store sensitive data directly in HDFS without encryption
D) Leverage Cloudera Knox Gateway for secure access to Spark applications
2. What mechanism does Airflow provide to retry failed tasks?
A) Manual intervention and rerun via the Airflow Webserver
B) The on failure callback function in DAG definitions
C) The retry_delay and retries parameters in task definitions
D) Airflow Scheduler's automatic rerun feature
3. You're working with a Spark application that processes sensitive dat
a. How can you ensure that persisted data remains secure even if accessed from unauthorized sources?
A) Encrypt the data before persisting it and decrypt it when needed
B) Implement custom access control mechanisms within your application
C) No additional security measures are needed, as Spark handles data security
D) Rely on Spark's lineage tracking to prevent unauthorized access
4. What is the significance of "Sort Merge Join" appearing in an Explain Plan in Cloudera's SQL engines?
A) It signifies that the join operation is performed by sorting and then merging two datasets, which can be efficient for large, sorted datasets
B) It suggests that the join operation is performed without sorting, leading to faster execution
C) It is the least preferred join method due to its high CPU usage
D) It indicates that the query will benefit from additional indexes
5. Why is it recommended to use the DataFrame API over RDDs for most data processing tasks in Spark?
A) RDDs are deprecated and will be removed in future versions of Spark.
B) DataFrames automatically optimize queries using the Catalyst optimizer and Tungsten execution engine.
C) DataFrames require less memory and compute resources compared to RDDs.
D) DataFrames provide more fine-grained control over partitioning and parallelism.
Fragen und Antworten:
| 1. Frage Antwort: A,D | 2. Frage Antwort: C | 3. Frage Antwort: A | 4. Frage Antwort: A | 5. Frage Antwort: B |
1535 Kundenbewertungen
