Amazon Redshift controls the workload on the console called WLM. One of the latest additions from Amazon is the support for automatic queue hopping. This is a great feature for users who want to run large queries and be able to handle the priority more intelligently.
Before releasing this functionality, users were required to stop slow running queries and change their query priorities by hand. It can sometimes be quite cumbersome and even impossible if they run thousands of queries per day. The automatic queue hopping feature allows them to manage these slow running queries by utilizing timeout settings.
When you set a timeout for a queue on the WLM panel in the Redshift console, the timeout for the queries in the queue are applied. If a query reaches the timeout period, it will automatically move it to another queue and won’t block any pending queries. In other words, when motorway traffic is backed up due to a stalled vehicle, the obstructing vehicle is removed and the traffic begins to flow again.
The logic for auto queue hopping is determined by the name of the Query Group or User Group.
This is the latest feature from Amazon that was announced on January 7, 2016.
Original Announcement