DAS-C01 · Question #174
An online retail company is using Amazon Redshift to run queries and perform analytics on customer shopping behavior. When multiple queries are running on the cluster, runtime for small queries…
The correct answer is C. Configure short query acceleration in workload management (WLM). Short Query Acceleration (SQA) in Amazon Redshift WLM is purpose-built for this scenario. SQA uses machine learning to predict a query's execution time and automatically routes queries predicted to be short-running to a dedicated, prioritized queue - bypassing longer-running…
Question
An online retail company is using Amazon Redshift to run queries and perform analytics on customer shopping behavior. When multiple queries are running on the cluster, runtime for small queries increases significantly. The company’s data analytics team to decrease the runtime of these small queries by prioritizing them ahead of large queries. Which solution will meet these requirements?
Options
- AUse Amazon Redshift Spectrum for small queries
- BIncrease the concurrency limit in workload management (WLM)
- CConfigure short query acceleration in workload management (WLM)
- DAdd a dedicated compute node for small queries
How the community answered
(60 responses)- A15% (9)
- B7% (4)
- C73% (44)
- D5% (3)
Explanation
Short Query Acceleration (SQA) in Amazon Redshift WLM is purpose-built for this scenario. SQA uses machine learning to predict a query's execution time and automatically routes queries predicted to be short-running to a dedicated, prioritized queue - bypassing longer-running queries. This directly solves the problem of small queries being delayed behind large ones. Using Redshift Spectrum (A) offloads queries to S3 but does not prioritize them. Increasing the WLM concurrency limit (B) allows more queries to run simultaneously but does not prioritize small ones. Adding a dedicated compute node (D) is not a valid Amazon Redshift configuration option.
Topics
Community Discussion
No community discussion yet for this question.