Google Professional Data Engineer

Google Professional Data Engineer

Get started today

Ultimate access to all questions.


You are tasked with creating a visualization for operations teams that will provide insights on network performance. The visualization must adhere to the following requirements:

  1. Data Scope: The telemetry data should encompass all 50,000 installations, covering the most recent 6 weeks with a sampling rate of once per minute.
  2. Timeliness: The report must be generated with no more than a 3-hour delay from live data.
  3. Relevance: The report should exclusively highlight suboptimal network links, ensuring that the most suboptimal links are prominently displayed at the top of the report.
  4. Regional Grouping and Filtering: Suboptimal links should be groupable and filterable by regional geography to allow for focused analysis.
  5. Performance: The report should load with a user response time of less than 5 seconds.
  6. Dynamic Visualizations: Create a data source that stores the last 6 weeks of data and supports dynamic visualizations, enabling viewers to examine various date ranges, specific geographic regions, and unique installation types.
  7. Consistency and Maintenance: Ensure that the visualizations always reflect the latest data without requiring modifications or updates to the visualizations every month.

Considering these requirements, how should you proceed with designing and implementing such a visualization?




Explanation:

The correct answer is B. Creating a small set of generalized charts and tables bound to criteria filters that allow value selection is the most efficient way to handle the ever-changing data without the need to update visualizations each month. This method ensures the visualizations remain dynamic, allowing users to filter and sort data in different ways as needed, without the overhead of predefining every possible combination of criteria. Options A and C involve creating multiple charts for each possible combination, which is impractical. Option D involves developing a complex application which is more time-consuming and possibly over-engineered for the given requirements.