ASSOCIATE-DATA-PRACTITIONER practice question 5 of 25

Your retail company collects customer data from various sources:Online transactions: Stored in a MySQL databaseCustomer feedback: Stored as text…

Full set: 106 questions

Your retail company collects customer data from various sources:

Online transactions: Stored in a MySQL database

Customer feedback: Stored as text files on a company server

Social media activity: Streamed in real-time from social media platforms

You are designing a data pipeline to extract this data. Which Google Cloud storage system(s) should you select for further analysis and ML model training?

Answer options
A:

A. 1. Online transactions: Cloud Storage 2. Customer feedback: Cloud Storage 3. Social media activity: Cloud Storage

B:

B. 1. Online transactions: BigQuery 2. Customer feedback: Cloud Storage 3. Social media activity: BigQuery

Correct answer
C:

C. 1. Online transactions: Bigtable 2. Customer feedback: Cloud Storage 3. Social media activity: CloudSQL for MySQL

D:

D. 1. Online transactions: Cloud SQL for MySQL 2. Customer feedback: BigQuery 3. Social media activity: Cloud Storage

Correct Answer:
B. B. 1. Online transactions: BigQuery 2. Customer feedback: Cloud Storage 3. Social media activity: BigQuery
Explanation

This selection aligns each data source with the optimal Google Cloud service for analysis and ML.
1. Online transactions (MySQL): Migrating this structured data to BigQuery, a serverless data warehouse, is ideal for large-scale SQL-based analysis and as a direct data source for ML model training.
2. Customer feedback (text files): Cloud Storage is the best choice for storing unstructured data like text files. It acts as a data lake where services like Dataproc or AI Platform can access the files for NLP tasks.
3. Social media activity (real-time): BigQuery excels at ingesting and analyzing high-volume, real-time streaming data, making it the perfect destination for social media feeds.

Why Incorrect

A. Storing structured transactional and streaming data in Cloud Storage is suboptimal for direct, high-performance querying and analysis compared to a dedicated data warehouse like BigQuery.
C. Bigtable is a NoSQL database for large analytical and operational workloads, not a relational data warehouse. Cloud SQL is not designed for high-throughput streaming ingestion.
D. Keeping transactions in Cloud SQL limits large-scale analytical capabilities. Storing raw text files in BigQuery is less conventional than using Cloud Storage as a data lake.

References

1. Google Cloud Documentation, "What is BigQuery?". It states, "BigQuery is a fully managed enterprise data warehouse that helps you manage and analyze your data... It's designed to handle petabyte-scale data and can ingest streaming data in real time."

2. Google Cloud Documentation, "What is Cloud Storage?". It describes Cloud Storage as the place for "storing and accessing unstructured data," making it suitable for a data lake.

3. Google Cloud Architecture Center, "Data warehouse modernization from on-premises to BigQuery". This guide outlines the pattern of moving transactional data to BigQuery for analytics.

Scroll to Top