Limit the total number of running instances launched from this AMI and associated configurations. Primarily useful to control the EC2 cost. Once the the instance number reaches the cap, the additional workload will sit in the queue and those projects will get less frequent builds.

Note that this cap is local to this specific AMI and associated configurations, and not to be confused with the global instance cap that you can specify at the cloud level, which affects the total number of instances this cloud will provision, across all the AMIs.