AWS RedShift: Concurrency scaling not adding clusters during spike

Question

Hello,   
  
We have set up Redshift and are trying to implement WorkLoadManagement. We have set out Concurrency Scaling mode to **auto** using manual and auto WLM. Tested both but Concurrency Scaling does not add additional  cluster during spike.   
  
Upon adding load onto the Redshift Cluster, the **CPU Utilisation** hits the 100% mark but upon checking the workload concurrency no additional clusters have been added.  
  
In my WLM Parameters, i have added max_concurrency_scaling_clsuters to 4. However, when the CPU Utilisation hits 100% for the compute nodes, a additional  clusters does not get added under Workload Concurrency

What could I be missing in my configuration that doe not allow an additional cluster to be added during a spike?  
  
  -- Link. to question asked on stackoverflow with attachments

Answer

Concurrency Scaling clusters are added when a WLM slots are taken by running queries and other queries are waiting for a WLM slot to execute.  
  
It _does not_ add clusters in response to high CPU since Redshift is designed to use as much CPU as possible (for maximum speed!).  
  
https://docs.aws.amazon.com/redshift/latest/dg/concurrency-scaling.html

AWS RedShift: Concurrency scaling not adding clusters during spike

Relevant content