Data Warehousing on AWS Training Cost Management 1 — Questions and Answers
Question 1: Which Amazon Redshift pricing model charges based on the number of queries run rather than cluster uptime?
- On-Demand pricing
- Reserved Instance pricing
- Redshift Serverless pricing (Correct answer)
- Spot Instance pricing
Correct answer: Redshift Serverless pricing
Redshift Serverless charges in Redshift Processing Units (RPUs) per second only when queries are running, with no idle costs.
Question 2: How do Reserved Instances reduce Amazon Redshift costs compared to On-Demand pricing?
- They provide automatic scaling at no extra charge
- They offer up to 75% discount in exchange for a 1- or 3-year commitment (Correct answer)
- They eliminate storage costs for the cluster
- They provide free data transfer between nodes
Correct answer: They offer up to 75% discount in exchange for a 1- or 3-year commitment
Reserved Instances provide significant discounts (up to 75%) over On-Demand rates when you commit to 1 or 3 years of usage.
Question 3: Which Redshift feature allows you to automatically scale compute capacity up or down based on workload demand?
- Manual Resize
- Elastic Resize
- Concurrency Scaling (Correct answer)
- Auto Scaling Groups
Correct answer: Concurrency Scaling
Concurrency Scaling automatically adds temporary cluster capacity during bursts of concurrent queries, charging only per second of use.
Question 4: What is the most cost-effective storage option for infrequently accessed historical data that still needs to be queryable via Redshift?
- Store in Redshift DC2 nodes
- Store in S3 and query with Redshift Spectrum (Correct answer)
- Store in Glacier and restore on demand
- Archive to EBS snapshots
Correct answer: Store in S3 and query with Redshift Spectrum
Redshift Spectrum queries data directly in S3 without loading it into the cluster, making S3 storage far cheaper than Redshift node storage.
Question 5: Which Redshift node type is recommended for large datasets (2+ TB per node) where storage cost is more important than raw compute speed?
- DC2.large
- DC2.8xlarge
- RA3.xlplus (Correct answer)
- DS2.8xlarge
Correct answer: RA3.xlplus
RA3 nodes use Amazon S3-backed Redshift Managed Storage, which is cheaper per TB than DC2 local SSD storage for large datasets.
Question 6: How can you reduce Redshift costs for a development cluster that is only used during business hours?
- Use Spot Instances for the cluster nodes
- Schedule cluster Pause and Resume to run only during work hours (Correct answer)
- Switch to Reserved Instances with partial upfront payment
- Enable Concurrency Scaling permanently
Correct answer: Schedule cluster Pause and Resume to run only during work hours
Scheduling cluster Pause during nights and weekends eliminates compute charges for ~65% of the week in a typical 9-to-5 scenario.
Data Warehousing on AWS Training Cost Management 1
Which Amazon Redshift pricing model charges based on the number of queries run rather than cluster uptime?