Billing dashboard overview in AWS - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ dax_lod_result

intermediate

Calculate Total Monthly Cost Using DAX

You have a billing dataset with columns: Date, Service, and Cost. Which DAX measure correctly calculates the total cost for the current month?

A Total Monthly Cost = CALCULATE(SUM(Billing[Cost]), FILTER(Billing, MONTH(Billing[Date]) = MONTH(TODAY()) && YEAR(Billing[Date]) = YEAR(TODAY())))

B Total Monthly Cost = SUMX(FILTER(Billing, Billing[Date] = TODAY()), Billing[Cost])

C Total Monthly Cost = CALCULATE(SUM(Billing[Cost]), ALL(Billing))

D Total Monthly Cost = SUM(Billing[Cost])

💡 Hint

Use CALCULATE with FILTER to restrict data to current month and year.

Option A correctly filters the billing data to only include rows where the date is in the current month and year, then sums the Cost.

Option A sums all costs without filtering by date.

Option A removes filters and sums all costs, ignoring date.

Option A filters only for today's date, not the whole month.

❓ visualization

intermediate

Best Visualization for Cost Trend Over Time

You want to show how AWS billing costs change month by month over the past year. Which visualization type is best suited for this?

A Bar chart showing total cost by service for the current month

B Pie chart showing cost distribution by service

C Line chart showing monthly total cost over time

D Scatter plot showing cost vs. number of resources

💡 Hint

Think about how to show changes over time clearly.

A line chart is ideal for showing trends over time, such as monthly billing costs.

Pie charts show parts of a whole at one point in time, not trends.

Bar charts can compare categories but are less effective for continuous time trends.

Scatter plots show relationships between two variables, not time trends.

❓ data_modeling

advanced

Designing a Data Model for Billing Analysis

You have raw AWS billing data with many services and usage types. To build an efficient billing dashboard, which data modeling approach is best?

A Store all raw data in a single flat table without relationships

B Create a star schema with a fact table for usage and dimension tables for services, dates, and accounts

C Use multiple unrelated tables for each service without common keys

D Create a snowflake schema with many normalized dimension tables but no fact table

💡 Hint

Think about how to organize data for fast queries and easy filtering.

A star schema with a fact table and dimension tables is best for BI dashboards because it supports efficient queries and clear relationships.

Option B leads to slow queries and difficulty filtering.

Option B lacks relationships, making analysis hard.

Option B is overly complex and less performant for typical BI needs.

🔧 Debug

advanced

Identify the Error in This DAX Measure

This DAX measure is intended to calculate the average daily cost for the current month but returns an error. What is the cause?

AWS

Avg Daily Cost = AVERAGEX(FILTER(Billing, MONTH(Billing[Date]) = MONTH(TODAY())), Billing[Cost])

A Billing[Cost] cannot be used inside AVERAGEX without aggregation

B AVERAGEX requires a table as the first argument, but FILTER returns a scalar

C TODAY() cannot be used inside FILTER

D The FILTER function is missing a YEAR condition, causing incorrect filtering

💡 Hint

Check if the filter includes the correct year to avoid mixing months from different years.

The measure filters only by month but not by year, so it mixes data from the same month in all years, causing wrong results or errors.

FILTER returns a table, so AVERAGEX works correctly.

Billing[Cost] is valid inside AVERAGEX.

TODAY() can be used inside FILTER.

🧠 Conceptual

expert

Scenario: Optimizing Billing Dashboard Performance

Your billing dashboard is slow because it queries a very large dataset live. Which approach will best improve performance without losing data accuracy?

A Implement incremental data refresh to load only new billing data daily and pre-aggregate monthly costs

B Use direct query mode to always get the latest data without caching

C Remove all filters and show raw data to speed up queries

D Split the dataset into many small unrelated tables to reduce size

💡 Hint

Think about balancing data freshness and query speed with pre-processing.

Incremental refresh loads only new data and pre-aggregations reduce query complexity, improving performance while keeping accuracy.

Removing filters shows too much data and slows queries.

Direct query mode can be slower on large datasets.

Splitting tables without relationships complicates analysis and may not improve speed.