Reputation: 5279
we have amazon redshift being used by ETL and by analyst. I would want to restrict all the analyst to be able to run the query for a maximum of 10 minutes, on the other hand I would want the queries run by the ETL task not to time out.
Using wlm seems a good idea. So I created two queues one for ETL task and one for analyst. But the wlm divides memory equally among the two queue. The ETL runs only during the night and the analyst run queries only during the day, do not want to waste memory this way
Upvotes: 0
Views: 731
Reputation: 800
WLM allows you to specify the percentage of memory that a queue gets: http://docs.aws.amazon.com/redshift/latest/dg/cm-c-defining-query-queues.html
Also, it allows you to dynamically change the allocation via API: http://docs.aws.amazon.com/cli/latest/reference/redshift/modify-cluster-parameter-group.html
For example, you can have Analyst queue: (90%, 5 slots), ETL queue: (10%, 1 slot) during the day, and Analyst queue: (10%, 1 slot), ETL queue: (90%, 1 slot) during the night.
Upvotes: 1