Cloudera CDP-3002 考試概覽:

認證廠商:	Cloudera
考試名稱:	CDP 資料工程師 - 認證考試
考試代碼:	CDP-3002
及格分數:	70%
證照有效期限:	2 年
考試時間:	120 minutes
考試形式:	選擇題, 實作實驗（表現導向題型）
支援語言:	English
實際考試題數:	60-70
考試費用:	USD 295
相關認證:	Cloudera Certified Associate (CCA) 資料分析師 Cloudera Certified Professional (CCP) 資料工程師
範例考題:	Cloudera CDP-3002 範例考題
考試方式:	可於授權考試中心應考，或採線上監考方式應考
必備條件:	建議具備：Cloudera CDP 實作經驗、熟悉 Python/Scala，並瞭解分散式資料處理概念
官方大綱網址:	https://www.cloudera.com/about/certification/cdp-certification.html

Cloudera CDP-3002 考試大綱主題:

章節	權重	目標
使用 Spark 進行資料處理	30%	- Spark 效能最佳化 - Spark 核心概念 - Spark 結構化串流 - DataFrame 與 Dataset 應用程式介面 - Spark SQL 與 DataFrames
CDP 平台營運管理	15%	- Cloudera 資料平台架構 - Cloudera 流程管理 - 資料湖與儲存系統 - 叢集管理與監控
資料品質與治理	15%	- 資料驗證與清理 - 資料溯源 - 存取控制與資安 - 資料目錄與中繼資料
資料擷取與整合	20%	- 資料轉換與 ETL - 串流資料擷取 - 批次資料擷取 - CDC (變更資料擷取) - 資料聯合
資料流程編排	20%	- 錯誤處理與重試機制 - CDP 上的 Apache Airflow - 工作流程相依性 - 流程排程與觸發條件

最新的 Cloudera Certification CDP-3002 免費考試真題：

1. You are working with a complex Spark application involving multiple stages, and you want to ensure that later stages only start processing after all data from the previous stage is complete. How can you achieve this dependency management in Spark?

A) Leverage Spark Streaming concepts like micro-batching for real-time data processing
B) Use explicit persist() calls with different storage levels for each stage
C) Rely on Spark's lineage tracking and stage boundaries to enforce dependencies
D) Implement custom logic with synchronization mechanisms between stages

2. In the context of big data processing, what is a potential downside of relying heavily on schema inference?

A) Enhanced data security and compliance
B) Increased data storage efficiency
C) Reduced flexibility in handling different data types
D) Potential performance overhead due to the dynamic analysis of data structure

3. How does Spark achieve fault tolerance during distributed processing?

A) By restarting failed tasks on different nodes
B) By replicating data across all nodes in the cluster
C) null
D) By implementing automatic checkpointing of intermediate results

4. Your team is integrating PySpark with a MySQL database. You need to read data from a table named 'employees'. Which of the following PySpark code snippets correctly accomplishes this task?

A)

B)

C)

D)

5. When monitoring a PySpark application in Kubernetes, you notice that Executor pods are frequently restarting. What is the most likely cause of this issue?

A) Too many tasks assigned to each Executor.
B) Executor pods are not configured to restart upon failure.
C) Insufficient memory allocation for Executor pods.
D) Dynamic resource allocation is disabled.

問題與答案：

問題 #1
答案： A,C

問題 #2
答案： D

問題 #3
答案： C

問題 #4
答案： D

問題 #5
答案： C