Amazon Redshift improved Short Query Acceleration (SQA) by automating the maximum time-out setting for short queries. On Nov 20th, 2017, we announced SQA which uses machine learning algorithms to predict the execution time of a query, and move short-running queries to an express queue for immediate processing. In the past, SQA needed user defined ‘maximum run time’ (between 1-20 seconds) to identify short queries. With this update, by setting the maximum short query run time to ‘dynamic,’ you can let Redshift automate this setting. In addition, SQA’s improved machine learning algorithm adapts the maximum run time to the changing workload, making it easier to use to minimize short query queuing time and increase throughput.

Source: Amazon Web Services News Amazon Redshift Makes Short Query Acceleration Self-Optimizing