Closed kfferrando closed 5 years ago
Nodes scaled to m5.xlarge. The reduced memory of the m5.large instances caused almost twice the nodes to be created based on our pod count so m5.xlarge is needed. We will continue to monitor performance and reduce pod counts if needed to have less nodes online.
As part of issue #40, scale down Frankfurt worker node sizes from m4.xlarge to m5.large when upgrading to the m5 class.